Load cooperative control and quick response method and system

Abstract

The invention discloses a load cooperative control and quick response method and system, and the method comprises the steps: classifying and recognizing different flexible load devices at a demand side, collecting multi-dimensional parameters, and analyzing the multi-time scale characteristics and nonlinear constraints of the multi-dimensional parameters to determine an operation boundary; a load precise adjustable capacity model is established by solving boundary vertexes of a real two-dimensional projection domain, a coupling relation is analyzed, and a cooperative control and unified scheduling model of the flexible load and the local power grid is established; a quick response control algorithm and a distributed resource online calculation optimization technology are designed, and a user side flexible load is incorporated into a provincial domain level virtual power plant standard system. The system comprises six units including a multi-dimensional acquisition unit, a characteristic analysis unit, a model construction unit, a collaborative scheduling unit, a response optimization unit and a standard incorporation unit which are connected in sequence to transmit data. According to the method and the system, the load control accuracy and the response speed are improved, the problems of inaccurate model and insufficient response in the prior art are effectively solved, and efficient and stable operation of a power system is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power system fast response technology, and in particular to a load coordination control and fast response method and system. Background Technology

[0002] As power systems evolve towards diversification and intelligence, the types of flexible loads on the demand side are constantly increasing, including large industrial loads, HVAC loads, charging loads, and energy storage devices. These loads exhibit multi-timescale variations during operation and interact with the power grid more frequently. Simultaneously, the construction of provincial-level virtual power plants is progressing, necessitating the inclusion of user-side flexible loads into a unified standard system for management to achieve efficient utilization of power resources and stable grid operation. In this context, traditional load control methods are no longer adequate to adapt to the complex and ever-changing load characteristics and the demands of large-scale coordinated dispatch. There is an urgent need to construct more precise and efficient load coordinated control and rapid response methods and systems to address the coordinated management of different types of flexible loads and their coordinated dispatch with the power grid, thereby improving the flexibility and response speed of load control and ensuring the safe and stable operation of the power system.

[0003] Existing technologies have two significant drawbacks in load coordination control and rapid response. Firstly, when constructing load adjustable capacity models and coordinated dispatch models, the nonlinear constraints of different types of flexible loads are not comprehensively considered, and the load operation boundaries that satisfy the constraints are not accurately solved. This results in the models failing to accurately reflect the actual adjustable capacity of various loads and their coordination with the power grid, thus affecting the accuracy and reliability of load dispatch and making it difficult to adapt to complex load operation scenarios. Secondly, in terms of rapid response control and distributed resource optimization, there is a lack of sound calibration calculation logic and parameter update mechanisms. The technologies fail to effectively combine the computing power, storage capacity, and network transmission speed of distributed resources to formulate reasonable resource allocation schemes, resulting in slow load response speeds. This makes it difficult to respond promptly to changes in the power grid's operating status, failing to meet the rapid load response requirements of provincial-level virtual power plants and impacting the overall operating efficiency of the power system. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a load coordination control and fast response method and system.

[0005] The technical solution adopted in this invention is:

[0006] A load coordination control and fast response method includes the following steps:

[0007] S1. Classify and identify flexible load equipment in different functional areas on the demand side, distinguishing between large industrial loads, HVAC loads, charging loads and energy storage equipment, and collect multi-dimensional parameter information during the operation of different types of equipment. The multi-dimensional parameter information includes equipment operating status parameters, load change parameters and parameters interacting with the power grid.

[0008] S2. Based on the collected multi-dimensional parameter information, analyze the multi-timescale characteristics of different types of flexible load equipment, identify the load variation law and nonlinear constraint conditions under different timescales, and determine the operating boundaries of different types of loads under different timescales.

[0009] S3. For different types of flexible load equipment, based on the analysis results of step S2, solve the boundary vertex of the real two-dimensional projection domain that satisfies the nonlinear state space constraint by translating the straight line corresponding to the characteristic normal vector, establish an accurate adjustable capacity model for different types of loads, and analyze the coupling relationship between different types of loads.

[0010] S4. Based on the established precise adjustable capacity model and coupling relationship, construct a collaborative control and unified dispatch model for flexible adjustable power load and local power grid, and clarify the parameter association rules in the collaborative control process;

[0011] S5. Based on the collaborative control and unified scheduling model, design a fast response control algorithm and a distributed resource online computing optimization technology, and determine the calibration calculation logic and parameter update mechanism in the algorithm and technology;

[0012] S6. Based on fast response control algorithms and distributed resource online computing optimization technology, user-side flexible loads are incorporated into the provincial-level virtual power plant standard system to achieve coordinated control and fast response of user-side flexible loads.

[0013] Furthermore, when establishing accurate adjustable capacity models for different types of loads in S3, the following model is used:

[0014] ,

[0015] in, Indicates precisely adjustable load capacity. Indicates the number of load devices. Indicates the first Adjustment coefficient of similar load equipment Indicates the first Power parameters of similar load equipment Indicates the first Operating time parameters of similar load equipment Indicates the first Status parameters of load-like equipment Represents the load adjustment function. Indicates a time interval.

[0016] Furthermore, when constructing the collaborative control and unified scheduling model in S4, the following model is adopted:

[0017] ,

[0018] in, This represents the target value for coordinated control of the local power grid. and Indicates the weighting coefficient. This indicates the number of types of flexible and adjustable power loads. Indicates the first Weighting parameters for flexible and adjustable power loads. Indicates the first Input parameters for flexible adjustable power loads, Indicates the first Output parameters of flexible adjustable power loads Represents the cooperative control function. These represent the operating constraint parameters of the local power grid.

[0019] Furthermore, in designing the fast response control algorithm in S5, the following algorithm model is adopted:

[0020] ,

[0021] in, Indicators representing rapid response speed This represents the load control objective function. Represents a time variable. Indicates the number of control parameters. Indicates the first One control input parameter, Indicates the first One control output parameter, Represents the response function. Represents the response coefficient. This represents the power deviation parameter.

[0022] Furthermore, the distributed resource online computing optimization technology in S5 adopts the following model:

[0023] ,in,

[0024] This indicates the online calculation optimization results. Indicates the number of distributed resources. Indicates the first Configuration parameters for a distributed resource. Indicates the first The operating parameters of a distributed resource. Represents the resource optimization function. Indicates the number of data samples. Indicates the first Each data sample value, Indicates the number of constraints. Indicates the first Constraint parameters.

[0025] Furthermore, when incorporating the provincial-level virtual power plant standard system into S6, the following model is adopted:

[0026] ,

[0027] in, Indicators representing the degree of conformity to the standard system. and This represents the proportionality coefficient. Indicates the number of indicators in the standard system. Indicates the first The actual value of each standard indicator Indicates the first The target value of each standard indicator This represents the function for calculating the fit. Indicates the number of evaluation dimensions. Indicates the first The score values ​​for each evaluation dimension.

[0028] Furthermore, step S3 includes the following sub-steps:

[0029] S31. Select nonlinear state-space constraint parameters for different types of flexible load equipment, determine the boundary range corresponding to each constraint parameter, and convert these constraint parameters into mathematical expressions.

[0030] S32. Calculate the characteristic normal vector corresponding to the mathematical expression of each constraint parameter. Based on the direction and magnitude of the normal vector, determine the direction and distance of the translation line to ensure that the translated line can accurately reflect the changes in the constraint conditions.

[0031] S33. Perform an intersection operation between the translated straight line and the two-dimensional projection domain, solve for the coordinate values ​​of the intersection points, filter these coordinate values, remove coordinate points that do not conform to the actual running conditions, and obtain the true boundary vertices of the two-dimensional projection domain.

[0032] S34. Based on the obtained boundary vertices and the operating parameters of different types of load equipment, construct an accurate adjustable capacity model. By comparing the model calculation results with the actual load adjustment data, adjust the parameters in the model so that the model reflects the adjustable capacity of the load.

[0033] Furthermore, step S4 includes the following sub-steps:

[0034] S41. Collect local power grid operation data, including power transmission data, voltage stability data and frequency fluctuation data, classify and organize this data, and extract data information related to flexible load collaborative control.

[0035] S42. Analyze the correlation between flexible and adjustable power load and local power grid operation data, identify the key parameters that affect the effect of coordinated control, and clarify the role of these key parameters in the coordinated control process;

[0036] S43. Based on key parameters and their relationships, design the framework structure of the collaborative control and unified scheduling model, determine the specific content of the input layer, processing layer and output layer of the model, and clarify the data transmission path between each layer.

[0037] S44. Input the organized running data into the model framework, process the data through the calculation logic of the model processing layer to obtain the collaborative control instructions, transmit the instructions to the output layer, and complete the construction of the collaborative control and unified scheduling model.

[0038] Furthermore, step S5 includes the following sub-steps:

[0039] S51. Determine the control objective of the fast response control algorithm, select the input parameters required by the algorithm around the objective, including load change rate, grid frequency deviation and power demand change parameters, and define the range of these parameters;

[0040] S52. Design the calibration calculation logic of the algorithm, including parameter comparison logic, instruction generation logic and deviation correction logic, and clarify the calculation order and data interaction method of each logic module;

[0041] S53. Collect relevant information about distributed resources, including the computing power, storage capacity and network transmission speed of the resources, and determine the resource allocation scheme of the online computing optimization technology for distributed resources based on this information;

[0042] S54. Combine the designed computational logic with the resource allocation scheme to construct a complete process for fast response control algorithm and distributed resource online computing optimization technology. Test each step in the process to ensure that the technology can operate normally.

[0043] A load coordination control and rapid response system, the system being applied to the aforementioned load coordination control and rapid response method, comprising:

[0044] The flexible load parameter multi-dimensional acquisition unit is used to collect the operating status parameters, load change parameters and grid interaction parameters of large industrial loads, HVAC loads, charging loads and energy storage equipment in different functional areas on the demand side, and transmit the collected parameters to the load characteristic analysis unit.

[0045] The load characteristic multi-timescale analysis unit receives parameters transmitted by the flexible load parameter multi-dimensional acquisition unit, analyzes the multi-timescale characteristics of different types of flexible load equipment, identifies the load variation law and nonlinear constraint conditions under different timescales, and determines the operating boundary of different types of loads under different timescales.

[0046] The load precise adjustable capacity model building unit receives the analysis results from the load characteristic multi-timescale analysis unit, solves the boundary vertices of the real two-dimensional projection domain that satisfy the nonlinear state space constraints by translating the straight line corresponding to the characteristic normal vector, establishes precise adjustable capacity models for different types of loads, analyzes the coupling relationship between different types of loads, and transmits the model to the collaborative scheduling model building unit.

[0047] The flexible load and grid coordinated dispatch model construction unit receives the model transmitted by the load precise adjustable capacity model construction unit, constructs a coordinated control and unified dispatch model of flexible adjustable power load and local power grid, clarifies the parameter association rules in the coordinated control process, and transmits the model to the response optimization technology design unit.

[0048] The rapid response and computational optimization technology design unit receives the model transmitted by the collaborative scheduling model construction unit, designs the rapid response control algorithm and the distributed resource online computational optimization technology, and determines the calibration computation logic and parameter update mechanism in the algorithm and technology;

[0049] The standard system of virtual power plants is incorporated into the execution unit. Based on the fast response control algorithm and distributed resource online computing optimization technology, the user-side flexible load is incorporated into the standard system of provincial-level virtual power plants to carry out coordinated control and fast response of the user-side flexible load. Different units are connected sequentially through data transmission lines to ensure the effective transmission of data and information.

[0050] The present invention has the following beneficial effects:

[0051] This invention proposes a load coordination control and rapid response method and system. By classifying and identifying different types of flexible loads on the demand side and collecting multi-dimensional parameters, and analyzing nonlinear constraints based on multi-timescale characteristics, it can accurately construct adjustable capacity models for various loads and coordinated dispatch models with the local power grid. Simultaneously, it designs a comprehensive rapid response control algorithm and distributed resource online computation optimization technology, and incorporates user-side flexible loads into the provincial-level virtual power plant standard system, effectively improving the accuracy, coordination, and response speed of load control, ensuring the efficient and stable operation of the power system. Addressing the inaccuracy of model construction in existing technologies, it solves for the boundary vertices of the real two-dimensional projection domain by translating the characteristic normal vectors, comprehensively considering nonlinear constraints, enabling the model to accurately reflect the load's adjustable capacity and its coordination with the power grid. Regarding the insufficient rapid response, it clarifies the calibration calculation logic and parameter update mechanism of the rapid response algorithm, and formulates optimization schemes based on distributed resource computing capabilities, storage capacity, and network transmission speed, significantly improving response efficiency and meeting the rapid load response requirements of provincial-level virtual power plants. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating the steps of a load coordination control and fast response method according to the present invention.

[0053] Figure 2 This is a unit composition diagram of a load coordination control and fast response system according to the present invention. Detailed Implementation

[0054] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] like Figure 1 As shown, a load coordination control and fast response method includes the following steps:

[0056] S1. Classify and identify flexible load equipment in different functional areas on the demand side, distinguishing between large industrial loads, HVAC loads, charging loads and energy storage equipment, and collect multi-dimensional parameter information during the operation of different types of equipment. The multi-dimensional parameter information includes equipment operating status parameters, load change parameters and parameters interacting with the power grid.

[0057] Specifically, in step S1, flexible load equipment in different functional areas such as industrial plants, commercial buildings, and residential communities on the demand side is first classified, clearly distinguishing between large industrial loads with a single unit power of 100 kW to 500 kW (such as motor equipment in production workshops), HVAC loads with a rated power of 5 kW to 30 kW (such as central air conditioning systems in commercial buildings), charging loads with an output power of 2 kW to 15 kW (such as electric vehicle charging piles in residential communities), and energy storage equipment with a capacity of 50 kWh to 500 kWh (such as lithium battery energy storage systems in factories). Then, sensors with a sampling frequency of 10 times per second are used to collect multi-dimensional parameter information during the operation of various types of equipment, including equipment operating status parameters. The data collection includes motor speed (ranging from 1000 rpm to 3000 rpm), air conditioning operation mode (cooling, heating, and ventilation), charging pile charging current (5 amps to 100 amps), and the state of charge of energy storage equipment (20% to 100%). Load variation parameters include hourly load power fluctuation (±5% to ±20%) and daily load peak occurrence time (8:00 to 10:00 and 18:00 to 20:00). Parameters interacting with the power grid include the voltage of the equipment connected to the grid (220 volts and 380 volts), power factor (0.8 to 0.95), and daily grid connection duration (8 hours to 24 hours). The data collection process lasts for 72 hours to obtain a complete sample of operating data, ensuring the comprehensiveness and representativeness of the parameter information.

[0058] S2. Based on the collected multi-dimensional parameter information, analyze the multi-timescale characteristics of different types of flexible load equipment, identify the load variation law and nonlinear constraint conditions under different timescales, and determine the operating boundaries of different types of loads under different timescales.

[0059] Specifically, during step S2, based on the multi-dimensional parameter information collected in S1 over 72 consecutive hours, data processing software is used to analyze the multi-timescale characteristics of various flexible load devices. Specifically, the ultra-short-term timescale (1 minute to 15 minutes) focuses on analyzing power surges in the charging load (e.g., the instantaneous power increase when a charging pile starts); the short-term timescale (1 hour to 6 hours) analyzes the pattern of HVAC load changes with ambient temperature (e.g., air conditioning power increases by 5% to 8% for every 1 degree Celsius increase in ambient temperature); and the medium- to long-term timescale (12 hours to 24 hours) analyzes the cyclical changes in large industrial loads (e.g., high load period from 8 AM to 8 PM, with power maintained at 300 kW to 500 kW, and low load period at other times, with power decreasing to 100 kW to 200 kW). Statistical analysis is used to identify load variation patterns at different time scales. For example, the standard deviation of power fluctuation for large industrial loads is 30 kW on weekdays and 15 kW on weekends. The peak daily load of HVAC loads is 20% to 30% higher in summer than in winter. Nonlinear constraints are defined, such as the charging and discharging power of energy storage devices cannot exceed 120% of the rated power (i.e., the charging and discharging power of a 50 kWh energy storage device cannot exceed 60 kW), and the total connected capacity of charging loads cannot exceed 30% of the capacity of the transformer in the same area (i.e., the total charging load capacity corresponding to a 500 kVA transformer cannot exceed 150 kVA). This allows for the determination of the operating boundaries of various loads at different time scales, such as the power operating range of large industrial loads from 100 kW to 500 kW and the operating temperature range of HVAC loads from 16 degrees Celsius to 28 degrees Celsius.

[0060] S3. For different types of flexible load equipment, by translating the straight line corresponding to the characteristic normal vector, solve the boundary vertex of the real two-dimensional projection domain that satisfies the nonlinear state space constraint, establish an accurate adjustable capacity model for different types of loads, and analyze the coupling relationship between different types of loads.

[0061] Specifically, in step S3, for different types of flexible load equipment, nonlinear state-space constraint parameters are first selected, such as motor torque (50 N·m to 200 N·m) for large industrial loads, heat exchanger efficiency (60% to 90%) for HVAC loads, battery temperature (0°C to 45°C) for charging loads, and charge / discharge rate (0.2C to 1C) for energy storage equipment. Then, the characteristic normal vector corresponding to each constraint parameter is solved using linear algebra calculation tools. For example, the normal vector corresponding to the motor torque constraint is along the positive direction of the torque variation axis and has a magnitude of 0.8. Based on the normal vector, the direction (in the same direction as the normal vector) and distance of the translation line are determined (the translation amount is 10% of the allowable fluctuation range of the constraint parameter; for example, if the allowable torque fluctuation is ±10 N·m, then the translation distance is 1 N·m). Subsequently, the translated line is compared with the two-dimensional projection domain (the horizontal axis is time). The intersection operation is performed on the vertical axis (load power). The coordinate values ​​of the intersection points are solved by coordinate calculation software. For example, after the large industrial load is translated into a straight line, the coordinates of the intersection with the projection domain are obtained as (300 kW at 8 o'clock) and (450 kW at 12 o'clock). These coordinate values ​​are filtered to remove coordinate points that exceed the actual power range of the equipment (such as below 100 kW or above 500 kW) to obtain the true two-dimensional projection domain boundary vertices. Finally, based on the boundary vertex data and combined with the operating parameters of various load equipment (such as motor speed and air conditioning power), an accurate adjustable capacity model is established. For example, the adjustable capacity range of the large industrial load is calculated to be 50 kW to 150 kW by the power difference corresponding to the boundary vertex. The coupling relationship between various loads is also analyzed. For example, when the power of the HVAC load increases by 10%, the discharge power of the energy storage equipment in the corresponding area needs to increase by 5% to maintain voltage stability.

[0062] S4. Based on the established precise adjustable capacity model and coupling relationship, construct a collaborative control and unified dispatch model for flexible adjustable power load and local power grid, and clarify the parameter association rules in the collaborative control process;

[0063] Specifically, during step S4, the operation data of the local power grid (covering an area of ​​5 to 20 square kilometers, including 10 to 30 distribution transformers) is first collected. This includes power transmission data (range of 5 MW to 20 MW), voltage stability data (220V ±7%, 380V ±7%), and frequency fluctuation data (50Hz ±0.2Hz) recorded every 15 minutes. This data is then correlated and matched with the precise adjustable capacity model data established in S3. Next, the coupling relationship between the adjustable capacity of various flexible loads and the power grid operation data is analyzed. For example, when the power transmission of the power grid is close to 20 MW (90% of the rated capacity), the adjustable capacity of the large industrial load needs to be reduced by 50 kW to 100 kW. When the voltage is below 205V, the adjustable discharge capacity of the energy storage device needs to be increased by 20 kW to 30 kW. Based on this, a collaborative control and unified scheduling model for flexible and adjustable power loads and local power grids is constructed. The model input layer includes adjustable capacity data for various loads and real-time power grid operation data. The processing layer adopts hierarchical control logic. The first layer classifies and schedules load types (prioritizing large industrial loads, followed by HVAC loads). The second layer adjusts load power according to the power grid frequency deviation (reducing load power by 5% to 10% when the frequency is below 49.8 Hz). The output layer generates specific load control instructions (such as "reduce the power of motors in workshop 3 by 30 kW" and "increase the discharge power of energy storage device 5 by 25 kW"). At the same time, the parameter association rules in the collaborative control process are clarified. For example, for every 0.1 Hz decrease in the power grid frequency, the total adjustable capacity of the loads called up needs to be increased by 2 MW to ensure that the model can adjust the load control strategy in real time according to the power grid status.

[0064] S5. Based on the collaborative control and unified scheduling model, design a fast response control algorithm and a distributed resource online computing optimization technology, and determine the calibration calculation logic and parameter update mechanism in the algorithm and technology;

[0065] Specifically, in step S5, based on the collaborative control and unified scheduling model constructed in S4, the control cycle of the fast response control algorithm is first determined to be 1 to 5 seconds to meet the grid's requirements for load response speed. The algorithm design adopts closed-loop control logic. The input parameters include the deviation between the real-time load power and the target power (range -50 kW to 50 kW) and the grid frequency change rate (range -0.1 Hz / s to 0.1 Hz / s). The control amount is calculated through proportional-integral-derivative control logic. For example, when the power deviation is 30 kW and the frequency change rate is -0.05 Hz / s, it is calculated that the load power needs to be reduced by 25 kW. At the same time, a parameter update mechanism is set up to adjust the proportional coefficient (range 0.5 to 2.0), integral time constant (range 1 to 10 seconds), and derivative time constant (range 0.1 to 1 second) in the algorithm every 30 minutes according to the grid operating status (such as total load and voltage level). For distributed resource online... The computation optimization technology first registers and manages computing resources distributed across different functional areas (such as 10 to 20 edge computing servers, each with a computing capacity of 20 billion to 50 billion operations per second). It then monitors the CPU utilization (range 20% to 80%), memory usage (range 30% to 70%), and network transmission rate (range 100Mbps to 1000Mbps) of each server in real time. Next, a resource allocation algorithm is designed. When there is a load-controlled computing task (data volume 50MB to 200MB, computation time requirement less than 1 second), the task is allocated based on the server's real-time status. For example, a task with 150MB of data is allocated to a server with 30% CPU utilization and a network transmission rate of 800Mbps, ensuring efficient completion of the computing task. Simultaneously, a task fault tolerance mechanism is established. When a server fails, the task is automatically switched to a backup server (the number of backup servers is 20% of the total number of servers).

[0066] S6. Based on fast response control algorithms and distributed resource online computing optimization technology, user-side flexible loads are incorporated into the provincial-level virtual power plant standard system to achieve coordinated control and fast response of user-side flexible loads.

[0067] Specifically, during step S6, the standard system requirements of the provincial-level virtual power plant are first obtained, including the communication protocol for load access (such as the IEC61850 protocol), data upload frequency (once every 1 to 5 minutes), load regulation response time (less than 10 seconds), and adjustable capacity declaration format (including equipment number, adjustable capacity range, and regulation period). Based on the fast response control algorithm and distributed resource online calculation optimization technology designed in S5, the control process of user-side flexible loads is modified. For example, real-time load data is encapsulated according to the IEC61850 protocol format to ensure that it is uploaded to the provincial-level virtual power plant platform every 3 minutes, and the response time of the fast response control algorithm is optimized to within 8 seconds to meet the response requirements of the virtual power plant. Subsequently, technical personnel are organized to debug the modified load equipment by simulating the regulation commands issued by the virtual power plant (such as "10 The system tests the load equipment's response accuracy (control error must be less than 5%) and timeliness (response delay less than 8 seconds) by reducing load power by 100 kW within minutes. After successful commissioning, each load equipment is assigned a unique provincial virtual power plant identification number. Finally, the load equipment's identification number, adjustable capacity range (e.g., 50 kW to 150 kW for large industrial loads, 30 kW to 80 kW for energy storage equipment), and control period (e.g., 8:00 to 22:00 on weekdays) are entered into the provincial virtual power plant management platform. This completes the process of incorporating user-side flexible loads into the provincial virtual power plant standard system, enabling coordinated control and rapid response of user-side flexible loads. For example, when the virtual power plant platform issues peak-shaving instructions, it can quickly allocate computing tasks through distributed resource online computing optimization technology and accurately control the power of various loads through rapid response control algorithms to ensure that the overall peak-shaving effect meets the requirements.

[0068] Preferably, when establishing accurate adjustable capacity models for different types of loads in S3, the following model is used: ,in, Indicates precisely adjustable load capacity. Indicates the number of load devices. Indicates the first Adjustment coefficient of similar load equipment Indicates the first Power parameters of similar load equipment Indicates the first Operating time parameters of similar load equipment Indicates the first Status parameters of load-like equipment Represents the load adjustment function. Indicates a time interval.

[0069] Specifically, in step S3, a precise adjustable capacity model for various types of loads is constructed. The calculation process of this model requires first determining the number of load devices. For scenarios such as industrial plants, commercial buildings, and residential communities, the number of similar load devices in a single scenario is usually between 10 and 50. Then, the adjustment coefficient for each type of load device is set. The value varies depending on the type of device. Large industrial loads have lower adjustment flexibility, so the adjustment coefficient is set to 0.6 to 0.8. HVAC loads have medium adjustment flexibility, so the adjustment coefficient is set to 0.8 to 0.9. Charging loads and energy storage devices have higher adjustment flexibility, so the adjustment coefficient is set to 0.9 to 0.95. Regarding power parameters, large industrial loads range from 100 kW to 500 kW, HVAC loads from 5 kW to 30 kW, and charging loads from 2 kW to 15 kW. Operating time parameters are determined based on the actual daily operating time of the equipment: large industrial loads typically operate for 8 to 16 hours, HVAC loads for 12 to 20 hours, charging loads for 6 to 12 hours, and energy storage devices for 24 hours. For status parameters, large industrial loads focus on operational stability, HVAC loads on operating modes, charging loads on the charging phase, and energy storage devices on state of charge. The load adjustment function is set according to equipment characteristics and is used to correlate power, time, status parameters, and adjustable capacity. The time interval is set from 15 to 60 minutes. The precise adjustable capacity of various loads is calculated through the model. This capacity value directly reflects the adjustment capability of different loads under specific conditions, providing data support for subsequent coordinated control. During implementation, parameters need to be adjusted based on actual equipment operating data to ensure that the calculation results are consistent with the actual adjustable capacity of the equipment, with the error controlled within 5%.

[0070] Preferably, when constructing the collaborative control and unified scheduling model in step S4, the following model is adopted: ,in, This represents the target value for coordinated control of the local power grid. and Indicates the weighting coefficient. This indicates the number of types of flexible and adjustable power loads. Indicates the first Weighting parameters for flexible and adjustable power loads. Indicates the first Input parameters for flexible adjustable power loads, Indicates the first Output parameters of flexible adjustable power loads Represents the cooperative control function. These represent the operating constraint parameters of the local power grid.

[0071] Specifically, step S4 constructs a collaborative control and unified dispatch model for flexible adjustable power loads and the local power grid, and calculates the collaborative control target value of the local power grid. The weighting coefficients are set according to the power grid's operational needs. When the power grid prioritizes load regulation accuracy, the weighting coefficient for the flexible adjustable power load is set to 0.6 to 0.8, and the weighting coefficient for the power grid's operational constraint parameters is set to 0.2 to 0.4. When the power grid prioritizes operational stability, the weighting coefficients are adjusted to 0.4 to 0.6 and 0.4 to 0.6, respectively. The types and number of flexible adjustable power loads are determined according to the actual load types participating in collaborative control, typically including four categories: large industrial loads, HVAC loads, charging loads, and energy storage devices. The weighting parameter for each type of load is set according to its proportion in the power grid. When the proportion of large industrial loads is high, the weighting parameter is 0.3 to 0.4; for HVAC loads, it is 0.2 to 0.3; and for charging loads and energy storage devices, it is 0.15 to 0.25 each. Input parameters include load power, operating time, and status data, while output parameters include power after load regulation and changes in operating status. The coordinated control function integrates these input and output parameters to establish the computational logic for load-grid coordination. Grid operating constraints include allowable voltage fluctuation ranges (220V ± 7%, 380V ± 7%), allowable frequency fluctuation ranges (50Hz ± 0.2Hz), and power transmission limits (determined by the grid's rated capacity, typically 20MW to 50MW). The model calculates the target value for coordinated control, which serves as the benchmark for grid coordinated control, guiding the generation of subsequent load regulation commands. During implementation, parameters need to be updated hourly to ensure the model adapts to real-time changes in the grid's operating status.

[0072] Preferably, when designing the fast response control algorithm in S5, the following algorithm model is adopted: ,in, Indicators representing rapid response speed This represents the load control objective function. Represents a time variable. Indicates the number of control parameters. Indicates the first One control input parameter, Indicates the first One control output parameter, Represents the response function. Represents the response coefficient. This represents the power deviation parameter.

[0073] Specifically, step S5 designs a fast response control algorithm and calculates the fast response speed index to evaluate the algorithm's response efficiency. The load regulation objective function is set around the grid's demand for load regulation, such as maintaining grid frequency stability and controlling power transmission within a reasonable range. The time variable is determined according to the algorithm's control cycle, typically 1 to 5 seconds. The number of control parameters is set according to the algorithm's complexity, including 8 to 12 parameters such as real-time load power, target power, frequency deviation, and power change rate. The input value of each control parameter needs to be collected in real time. Input parameters, such as real-time load power, are taken as the actual operating power of the current equipment, while output parameters, such as the adjusted target power, are calculated based on the grid's demand. The response function is used to correlate the input and output parameters with the response speed, reflecting the degree of influence of parameter changes on the response speed. The response coefficient is set according to the grid's requirements for response speed. When the grid requires a fast response, the coefficient is set to 0.8 to 1.0; when the grid allows for a moderate delay, the coefficient is set to 0.5 to 0.8. The power deviation parameter is the difference between the real-time load power and the target power, with an allowable range of -50 kW to 50 kW. The fast response speed index is calculated by model and the index value needs to be controlled between 0.5 seconds and 3 seconds to ensure that the algorithm can respond to the power grid control needs in a timely manner. During implementation, the index value needs to be monitored every 30 minutes. If it exceeds the reasonable range, the response coefficient and control parameters should be adjusted to optimize the algorithm response performance.

[0074] Preferably, the distributed resource online computing optimization technology in S5 adopts the following model: ,in, This indicates the online calculation optimization results. Indicates the number of distributed resources. Indicates the first Configuration parameters for a distributed resource. Indicates the first The operating parameters of a distributed resource. Represents the resource optimization function. Indicates the number of data samples. Indicates the first Each data sample value, Indicates the number of constraints. Indicates the first Constraint parameters.

[0075] Specifically, in step S5, the distributed resource online computing optimization technology calculates the online computing optimization results to evaluate the rationality of resource allocation. The number of distributed resources is determined according to the size of the coverage area: 10 to 20 edge computing servers are configured for small areas, 20 to 30 for medium-sized areas, and 30 to 50 for large areas. The configuration parameters of each distributed resource include CPU model, memory capacity, storage capacity, etc., and the operating parameters include CPU utilization, memory usage, network transmission rate, etc. CPU utilization needs to be controlled between 20% and 80%, memory usage needs to be controlled between 30% and 70%, and network transmission rate needs to be maintained between 100Mbps and 1000Mbps. The resource optimization function is designed based on resource configuration and operating parameters to calculate the optimization efficiency of a single resource. The number of data samples is determined by the daily computing workload, typically ranging from 1000 to 5000. Each data sample value includes information such as task data volume, computation time, and resource usage. The number of constraints includes 5 to 8 conditions such as computation time limits (less than 1 second), resource usage limits, and task priorities. Each constraint parameter is set according to actual needs, such as setting the computation time limit parameter to 1 second. The online computation optimization results are obtained through model calculations. The results must reflect the balance and efficiency of resource allocation, ensuring that more than 95% of the computing tasks can be completed within the specified time. During implementation, resource operating parameters need to be monitored in real time, and the resource allocation scheme needs to be dynamically adjusted to optimize the calculation results.

[0076] Preferably, when incorporating the provincial-level virtual power plant standard system in step S6, the following model is adopted: ,in, Indicators representing the degree of conformity to the standard system. and This represents the proportionality coefficient. Indicates the number of indicators in the standard system. Indicates the first The actual value of each standard indicator Indicates the first The target value of each standard indicator This represents the function for calculating the fit. Indicates the number of evaluation dimensions. Indicates the first The score values ​​for each evaluation dimension.

[0077] Specifically, step S6 incorporates user-side flexible loads into the provincial-level virtual power plant standard system, calculating the conformity index to determine feasibility of inclusion. The proportional coefficient is set according to the focus of the standard system. When the focus is on conformity to standard indicators, the proportional coefficient for the corresponding standard system indicator is set to 0.6 to 0.8, and the proportional coefficient for the corresponding evaluation dimension is set to 0.2 to 0.4. When the focus is on comprehensive evaluation, both proportional coefficients are adjusted to 0.5 to 0.5. The number of standard system indicators typically includes 8 to 12, such as communication protocols, data upload frequency, response time, and adjustable capacity declaration. The actual value of each indicator is collected through detection equipment. For example, the actual value of the data upload frequency is obtained through monitoring data transmission records. The target value is set according to the virtual power plant requirements, such as a target value of once every 1 to 5 minutes for the data upload frequency. The conformity calculation function is used to compare the actual value with the target value to calculate the conformity of a single indicator. The evaluation dimensions include 5 to 8 aspects such as technical compatibility, control reliability, and data security. The score for each evaluation dimension is obtained through expert review or automated testing, with a maximum score of 100 points and a passing score of 80 points. A conformity index is calculated through a model, and the index value must reach 80 points or above to be included in the virtual power plant standard system. During implementation, the index needs to be recalculated every quarter. If the index is lower than the passing score, the load equipment or control process needs to be modified until the standard requirements are met.

[0078] Preferably, step S3 includes the following sub-steps: S31, selecting nonlinear state-space constraint parameters for different types of flexible load equipment, determining the boundary range corresponding to each constraint parameter, and converting these constraint parameters into mathematical expressions; S32, calculating the characteristic normal vector corresponding to the mathematical expression of each constraint parameter, and determining the direction and distance of the translation line based on the direction and magnitude of the normal vector to ensure that the translated line can accurately reflect the changes in constraint conditions; S33, performing an intersection operation between the translated line and the two-dimensional projection domain, solving for the coordinate values ​​of the intersection points, filtering these coordinate values, removing coordinate points that do not conform to the actual operating conditions, and obtaining the true boundary vertices of the two-dimensional projection domain; S34, based on the obtained boundary vertices and combined with the operating parameters of different types of load equipment, constructing an accurate adjustable capacity model, and adjusting the parameters in the model by comparing the model calculation results with the actual load adjustment data to make the model reflect the adjustable capacity of the load.

[0079] Specifically, step S3 includes four sub-steps. In S31, nonlinear state-space constraint parameters for various flexible load equipment are first selected, such as motor torque (50 N·m to 200 N·m) for large industrial loads and heat exchanger efficiency (60% to 90%) for HVAC loads. After clarifying the boundary range of each parameter, it is transformed into a mathematical expression that includes the parameter value range and associated conditions, ensuring that the expression can accurately reflect the constraint logic of the parameter on the load operation. In S32, the characteristic normal vector corresponding to the mathematical expression of each constraint parameter is calculated using linear algebra tools. Based on the direction of the normal vector (e.g., along the torque growth direction) and magnitude (e.g., a vector magnitude of 0.8), the direction (same as the normal vector) and distance of the translation line are determined (take 10% of the allowable fluctuation range of the constraint parameter, such as 1 N·m for torque allowance ±10 N·m), ensuring that the translation line can accurately reflect the load operation. The mapping constraints change; in S33, a coordinate calculation tool is used to perform an intersection operation between the translated straight line and the two-dimensional projection domain (horizontal axis is time, vertical axis is load power), and the coordinate values ​​of the intersection point are obtained. Then, the coordinate points are selected according to the actual operating power range of the equipment (e.g., 100 kW to 500 kW for large industrial loads), and invalid points that are out of range are eliminated to obtain the true boundary vertices of the two-dimensional projection domain; in S34, based on the boundary vertex data and combined with the real-time operating parameters of the equipment (e.g., motor speed, air conditioning power), an accurate adjustable capacity model is constructed. By comparing the adjustable capacity value calculated by the model with the actual load adjustment data (e.g., 50 kW to 150 kW for large industrial loads), the coefficient parameters in the model are adjusted to keep the model calculation error within 5%, ensuring that the model can accurately reflect the load adjustable capacity and provide reliable load adjustment data support for subsequent coordinated control.

[0080] Preferably, step S4 includes the following sub-steps: S41, collecting local power grid operation data, including power transmission data, voltage stability data, and frequency fluctuation data; classifying and organizing this data; and extracting data information related to flexible load collaborative control; S42, analyzing the correlation between flexible adjustable power load and local power grid operation data; determining key parameters affecting the collaborative control effect; and clarifying the role of these key parameters in the collaborative control process; S43, based on the key parameters and correlations, designing the framework structure of the collaborative control and unified scheduling model; determining the specific content of the model's input layer, processing layer, and output layer; and clarifying the data transmission paths between each layer; S44, inputting the organized operation data into the model framework; processing the data through the computational logic of the model's processing layer to obtain collaborative control commands; transmitting the commands to the output layer; and completing the construction of the collaborative control and unified scheduling model.

[0081] Specifically, step S4 includes four sub-steps. In S41, operational data from the local power grid (covering 5 to 20 square kilometers) is collected, including power transmission data (5 MW to 20 MW) recorded every 15 minutes, voltage stability data (220V ± 7%, 380V ± 7%), and frequency fluctuation data (50Hz ± 0.2Hz). Data is then organized using data classification software according to the dimensions of "power grid parameter type - acquisition time - corresponding area," extracting data related to flexible load collaborative control (such as the correlation data between load power changes and power grid frequency fluctuations). In S42, data correlation analysis tools are used to calculate the correlation coefficient between flexible adjustable power load parameters (such as adjustable capacity and power change rate) and power grid operational data, determining key parameters affecting the collaborative control effect (such as power grid frequency deviation and total adjustable load capacity), and clarifying the role of these key parameters in collaborative control. Whether it's the trigger condition or the basis for adjustment; S43, based on key parameters and their correlations, designs a three-layer framework for the collaborative control and unified scheduling model. The input layer is set as adjustable load capacity data and real-time grid operation data; the processing layer is planned as a parameter comparison module, an instruction generation module, and a deviation correction module; and the output layer is determined as the load control instruction. At the same time, the data transmission path between each layer is clearly defined (e.g., input layer data is transmitted to the processing layer after verification, and the calculation results of the processing layer are transmitted to the output layer after review). In S44, the processed grid operation data is input into the model input layer. The parameter comparison module of the processing layer compares the actual value with the threshold (e.g., the frequency threshold from 49.8 Hz to 50.2 Hz). The instruction generation module generates control instructions based on the comparison results. After the deviation correction module corrects the instruction deviation, the instruction is transmitted to the output layer, completing the model construction and ensuring that the model can output accurate collaborative control instructions according to the grid status.

[0082] Preferably, step S5 includes the following sub-steps: S51, determining the control objective of the fast response control algorithm, selecting the input parameters required by the algorithm around the objective, including load change rate, grid frequency deviation, and power demand change parameters, and defining the range of these parameters; S52, designing the calibration calculation logic of the algorithm, including parameter comparison logic, instruction generation logic, and deviation correction logic, and clarifying the calculation order and data interaction method of each logic module; S53, collecting relevant information of distributed resources, including the computing power, storage capacity, and network transmission speed of the resources, and determining the resource allocation scheme of the distributed resource online computing optimization technology based on this information; S54, combining the designed calculation logic with the resource allocation scheme to construct a complete process of the fast response control algorithm and the distributed resource online computing optimization technology, and testing each link in the process to ensure that the technology can operate normally.

[0083] Specifically, step S5 includes four sub-steps. In S51, the control objective of the fast response control algorithm is first determined (e.g., controlling grid frequency fluctuations within ±0.1 Hz). Input parameters are selected around this objective, including load change rate (±5% / min to ±20% / min), grid frequency deviation (-0.2 Hz to 0.2 Hz), and power demand change (-5 MW to 5 MW). The effective value range of each parameter is defined using a parameter calibration tool to avoid invalid parameters affecting the algorithm calculation. In S52, the algorithm calibration calculation logic is designed. The parameter comparison logic uses a threshold comparison method (e.g., comparing the actual frequency with thresholds of 49.8 Hz and 50.2 Hz). The instruction generation logic generates preliminary instructions based on the comparison results according to the rule of "reducing load when the frequency is low and increasing load when the frequency is high". The deviation correction logic corrects the instruction quantity by comparing the parameter changes before and after the instruction execution (e.g., whether the frequency rebounds after executing the load reduction instruction). At the same time, the calculation order of each logic module (parameter comparison → instruction generation → deviation correction) and the number of... According to the interaction method (the calculation results of the previous module are directly transmitted to the next module); S53 collects distributed resource information, including the computing power (20 billion to 50 billion times / second), storage capacity (500GB to 2TB), and network transmission speed (100Mbps to 1000Mbps) of 10 to 30 edge computing servers. Based on this information, a resource allocation plan is formulated using a resource assessment tool (e.g., high-data-volume tasks are allocated to servers with a computing power ≥30 billion times / second). In S54, the calculation logic is combined with the resource allocation plan to build a complete process of "parameter input → logic calculation → resource allocation → task execution". The process testing tool simulates different power grid scenarios (e.g., frequency drop, power surge) and records the time consumption of each link (the total time consumption is required to be <5 seconds). For links with excessive time consumption (e.g., resource allocation time >1 second), the algorithm is optimized to ensure that the fast response control algorithm and distributed resource online calculation optimization technology can operate efficiently and meet the needs of rapid power grid regulation.

[0084] like Figure 2 As shown, a load coordination control and fast response system is applied to a load coordination control and fast response method, comprising:

[0085] The flexible load parameter multi-dimensional acquisition unit is used to collect the operating status parameters, load change parameters and grid interaction parameters of large industrial loads, HVAC loads, charging loads and energy storage equipment in different functional areas on the demand side, and transmit the collected parameters to the load characteristic analysis unit.

[0086] The load characteristic multi-timescale analysis unit receives parameters transmitted from the flexible load parameter multi-dimensional acquisition unit, analyzes the multi-timescale characteristics and nonlinear constraints of different types of flexible load equipment, determines the operating boundary, and transmits the analysis results to the adjustable capacity model construction unit.

[0087] The load precise adjustable capacity model construction unit receives the analysis results from the load characteristic multi-timescale analysis unit, solves the boundary vertices of the real two-dimensional projection domain by the straight line corresponding to the translation characteristic normal vector, establishes a precise adjustable capacity model and analyzes the coupling relationship, and transmits the model to the collaborative scheduling model construction unit.

[0088] The flexible load and grid coordinated dispatch model construction unit receives the model transmitted by the load precise adjustable capacity model construction unit, constructs a coordinated control and unified dispatch model of flexible adjustable power load and local power grid, clarifies parameter association rules, and transmits the model to the response optimization technology design unit.

[0089] The rapid response and computational optimization technology design unit receives the model transmitted by the collaborative scheduling model construction unit, designs the rapid response control algorithm and distributed resource online computational optimization technology, determines the calibration calculation logic and parameter update mechanism, and transmits the technology to the virtual power plant standard incorporation unit.

[0090] The virtual power plant standard system is incorporated into the execution unit, receiving the technology transmitted by the rapid response and computational optimization technology design unit. User-side flexible loads are incorporated into the provincial-level virtual power plant standard system for coordinated control and rapid response. Different units are connected sequentially through data transmission lines to ensure the effective transmission of data and information.

[0091] At the load management level, this invention can accurately classify and identify large industrial loads, HVAC loads, charging loads, and energy storage devices in different functional areas on the demand side, and comprehensively collect multi-dimensional operating parameters to provide detailed data support for subsequent analysis and control. At the model building level, it can combine the multi-timescale characteristics of loads and nonlinear constraints to establish an accurate load adjustable capacity model and a collaborative scheduling model of flexible loads and local power grids, ensuring that the model can truly reflect the load characteristics and the interaction between the load and the power grid. At the response and optimization level, the designed fast response control algorithm and distributed resource online calculation and optimization technology can efficiently handle load regulation needs, while incorporating user-side flexible loads into the provincial-level virtual power plant standard system to achieve large-scale, high-standard load collaborative management.

[0092] To address the issue of inaccurate model construction, this method and system solves the problem of the boundary vertices of the real two-dimensional projection domain that satisfy nonlinear state-space constraints by translating the straight line corresponding to the characteristic normal vector when establishing the load adjustable capacity model. This fully considers the nonlinear constraints of various loads and adjusts the model parameters to match the actual load adjustment data, significantly improving the model accuracy. Regarding the issue of insufficient fast response, the method clearly defines the calculation logic and parameter update mechanism when designing the fast response control algorithm. It also develops an optimized resource allocation scheme by combining the computing power, storage capacity, and network transmission speed of distributed resources, and constructs a complete response and calculation process to ensure timely response to changes in the power grid operating status and meet the needs of provincial-level virtual power plants for fast load response.

[0093] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A load coordination control and fast response method, characterized in that, Includes the following steps: S1. Classify and identify flexible load equipment in different functional areas on the demand side, distinguishing between large industrial loads, HVAC loads, charging loads and energy storage equipment, and collect multi-dimensional parameter information during the operation of different types of equipment. The multi-dimensional parameter information includes equipment operating status parameters, load change parameters and parameters interacting with the power grid. S2. Based on the collected multi-dimensional parameter information, analyze the multi-timescale characteristics of different types of flexible load equipment, identify the load variation law and nonlinear constraint conditions under different timescales, and determine the operating boundaries of different types of loads under different timescales. S3. For different types of flexible load equipment, based on the analysis results of step S2, solve the boundary vertex of the real two-dimensional projection domain that satisfies the nonlinear state space constraint by translating the straight line corresponding to the characteristic normal vector, establish an accurate adjustable capacity model for different types of loads, and analyze the coupling relationship between different types of loads. S4. Based on the established precise adjustable capacity model and coupling relationship, construct a collaborative control and unified dispatch model for flexible adjustable power load and local power grid, and clarify the parameter association rules in the collaborative control process; S5. Based on the collaborative control and unified scheduling model, design a fast response control algorithm and a distributed resource online computing optimization technology, and determine the calibration calculation logic and parameter update mechanism in the algorithm and technology; S6. Based on fast response control algorithms and distributed resource online computing optimization technology, user-side flexible loads are incorporated into the provincial-level virtual power plant standard system to achieve coordinated control and fast response of user-side flexible loads.

2. The load coordination control and fast response method according to claim 1, characterized in that, When establishing accurate adjustable capacity models for different types of loads in S3, the following model is used: ， in, Indicates precisely adjustable load capacity. Indicates the number of load devices. Indicates the first Adjustment coefficient of similar load equipment Indicates the first Power parameters of similar load equipment Indicates the first Operating time parameters of similar load equipment Indicates the first Status parameters of load-like equipment Represents the load adjustment function. Indicates a time interval.

3. The load coordination control and fast response method according to claim 1, characterized in that, When constructing the collaborative control and unified scheduling model in S4, the following model is adopted: ， in, This represents the target value for coordinated control of the local power grid. and Indicates the weighting coefficient. This indicates the number of types of flexible and adjustable power loads. Indicates the first Weighting parameters for flexible and adjustable power loads. Indicates the first Input parameters for flexible adjustable power loads, Indicates the first Output parameters of flexible adjustable power loads Represents the cooperative control function. These represent the operating constraint parameters of the local power grid.

4. The load coordination control and fast response method according to claim 1, characterized in that, When designing the fast response control algorithm in S5, the following algorithm model is adopted: ， in, Indicators representing rapid response speed This represents the load control objective function. Represents a time variable. Indicates the number of control parameters. Indicates the first One control input parameter, Indicates the first One control output parameter, Represents the response function. Represents the response coefficient. This represents the power deviation parameter.

5. The load coordination control and fast response method according to claim 1, characterized in that, The distributed resource online computation optimization technology in S5 adopts the following model: ,in, This indicates the online calculation optimization results. Indicates the number of distributed resources. Indicates the first Configuration parameters for a distributed resource. Indicates the first The operating parameters of a distributed resource. Represents the resource optimization function. Indicates the number of data samples. Indicates the first Each data sample value, Indicates the number of constraints. Indicates the first Constraint parameters.

6. The load coordination control and fast response method according to claim 1, characterized in that, When incorporating the provincial-level virtual power plant standard system into S6, the following model is adopted: ， in, Indicators representing the degree of conformity to the standard system. and This represents the proportionality coefficient. Indicates the number of indicators in the standard system. Indicates the first The actual value of each standard indicator Indicates the first The target value of each standard indicator This represents the function for calculating the fit. Indicates the number of evaluation dimensions. Indicates the first The score values ​​for each evaluation dimension.

7. The load coordination control and fast response method according to claim 1, characterized in that, S3 includes the following steps: S31. Select nonlinear state-space constraint parameters for different types of flexible load equipment, determine the boundary range corresponding to each constraint parameter, and convert these constraint parameters into mathematical expressions. S32. Calculate the characteristic normal vector corresponding to the mathematical expression of each constraint parameter. Based on the direction and magnitude of the normal vector, determine the direction and distance of the translation line to ensure that the translated line can accurately reflect the changes in the constraint conditions. S33. Perform an intersection operation between the translated straight line and the two-dimensional projection domain, solve for the coordinate values ​​of the intersection points, filter these coordinate values, remove coordinate points that do not conform to the actual running conditions, and obtain the true boundary vertices of the two-dimensional projection domain. S34. Based on the obtained boundary vertices and the operating parameters of different types of load equipment, construct an accurate adjustable capacity model. By comparing the model calculation results with the actual load adjustment data, adjust the parameters in the model so that the model reflects the adjustable capacity of the load.

8. The load coordination control and fast response method according to claim 1, characterized in that, S4 includes the following steps: S41. Collect local power grid operation data, including power transmission data, voltage stability data and frequency fluctuation data, classify and organize this data, and extract data information related to flexible load collaborative control. S42. Analyze the correlation between flexible and adjustable power load and local power grid operation data, identify the key parameters that affect the effect of coordinated control, and clarify the role of these key parameters in the coordinated control process; S43. Based on key parameters and their relationships, design the framework structure of the collaborative control and unified scheduling model, determine the specific content of the input layer, processing layer and output layer of the model, and clarify the data transmission path between each layer. S44. Input the organized running data into the model framework, process the data through the calculation logic of the model processing layer to obtain the collaborative control instructions, transmit the instructions to the output layer, and complete the construction of the collaborative control and unified scheduling model.

9. The load coordination control and fast response method according to claim 1, characterized in that, S5 includes the following steps: S51. Determine the control objective of the fast response control algorithm, select the input parameters required by the algorithm around the objective, including load change rate, grid frequency deviation and power demand change parameters, and define the range of these parameters; S52. Design the calibration calculation logic of the algorithm, including parameter comparison logic, instruction generation logic and deviation correction logic, and clarify the calculation order and data interaction method of each logic module; S53. Collect relevant information about distributed resources, including the computing power, storage capacity and network transmission speed of the resources, and determine the resource allocation scheme of the online computing optimization technology for distributed resources based on this information; S54. Combine the designed computational logic with the resource allocation scheme to construct a complete process for fast response control algorithm and distributed resource online computing optimization technology. Test each step in the process to ensure that the technology can operate normally.

10. A load coordination control and fast response system, characterized in that, The system is applied to the load coordination control and fast response method described in claim 1, comprising: The flexible load parameter multi-dimensional acquisition unit is used to collect the operating status parameters, load change parameters and grid interaction parameters of large industrial loads, HVAC loads, charging loads and energy storage equipment in different functional areas on the demand side, and transmit the collected parameters to the load characteristic analysis unit. The load characteristic multi-timescale analysis unit receives parameters transmitted by the flexible load parameter multi-dimensional acquisition unit, analyzes the multi-timescale characteristics of different types of flexible load equipment, identifies the load variation law and nonlinear constraint conditions under different timescales, and determines the operating boundary of different types of loads under different timescales. The load precise adjustable capacity model building unit receives the analysis results from the load characteristic multi-timescale analysis unit, solves the boundary vertices of the real two-dimensional projection domain that satisfy the nonlinear state space constraints by translating the straight line corresponding to the characteristic normal vector, establishes precise adjustable capacity models for different types of loads, analyzes the coupling relationship between different types of loads, and transmits the model to the collaborative scheduling model building unit. The flexible load and grid coordinated dispatch model construction unit receives the model transmitted by the load precise adjustable capacity model construction unit, constructs a coordinated control and unified dispatch model of flexible adjustable power load and local power grid, clarifies the parameter association rules in the coordinated control process, and transmits the model to the response optimization technology design unit. The rapid response and computational optimization technology design unit receives the model transmitted by the collaborative scheduling model construction unit, designs the rapid response control algorithm and the distributed resource online computational optimization technology, and determines the calibration computation logic and parameter update mechanism in the algorithm and technology; The standard system of virtual power plants is incorporated into the execution unit. Based on the fast response control algorithm and distributed resource online computing optimization technology, the user-side flexible load is incorporated into the standard system of provincial-level virtual power plants to carry out coordinated control and fast response of the user-side flexible load. Different units are connected sequentially through data transmission lines to ensure the effective transmission of data and information.

Images