Multi-energy collaborative scheduling method for new energy base based on wind-solar-storage coordination

Abstract

This invention discloses a multi-energy coordinated scheduling method for new energy bases based on wind, solar, and energy storage coordination, relating to the field of power system technology. It includes: constructing a hierarchical three-level coordinated architecture of base layer, power station layer, and equipment layer, achieving the integration of multi-timescale rolling correction and event-triggered coordination. Specifically, the base layer energy management platform generates baseline planning curves and voltage reference values; the power station layer edge coordinated controller receives real-time data to dynamically correct power station output commands and executes scheduling decisions based on abnormal event-triggered coordinated response logic; and the equipment layer execution unit achieves rapid power sharing based on a high-speed communication network and triggers an autonomous response mechanism when disturbed. This invention reduces the reliance on high-precision prediction and complex optimization, achieves spatiotemporal multi-energy complementarity and rapid response under sudden operating conditions, significantly improving the robustness, economy, and equipment lifespan of new energy base operations.

Description

Technical Field

[0001] This invention relates to the field of power systems, and specifically to a multi-energy coordinated dispatch method for new energy bases based on wind, solar and energy storage coordination. Background Technology

[0002] With the deepening of global energy structure transformation and the "dual carbon" goal, new energy power generation technologies, represented by wind power and photovoltaics, are developing rapidly and their penetration rate in the power system is continuously increasing. Wind and photovoltaic resources are widely distributed, clean, and renewable; their large-scale development and utilization are of significant strategic importance for optimizing the energy structure, reducing carbon emissions, and ensuring energy supply security. In the "Three Norths" region (Northwest, North, and Northwest China) and large-scale renewable energy demonstration bases, where new energy resources are abundant, the installed capacity of wind and photovoltaic power is growing rapidly, forming large-scale centralized new energy power generation clusters. However, wind and photovoltaic power generation has inherent intermittency and volatility; their output characteristics are significantly affected by meteorological conditions, exhibiting randomness, intermittency, and anti-peak-shaving characteristics, posing severe challenges to the peak-shaving, frequency regulation, and stable operation of the power system. Energy storage technologies, especially lithium-ion battery energy storage, flow battery energy storage, and pumped hydro storage, possess rapid response and energy spatiotemporal migration capabilities, making them key technologies for smoothing new energy fluctuations and improving system flexibility and dispatchability. Integrating energy storage with wind power and photovoltaics to build a coordinated wind-solar-storage integrated operation system has become an inevitable choice and industry consensus for improving the operational efficiency of new energy bases, ensuring the safety and stability of the power grid, and promoting the consumption of new energy.

[0003] Among them, wind-solar-storage coordinated dispatch technology is the core technical means to achieve efficient, economical, and safe operation of new energy bases. This technology aims to fully leverage the advantages of multi-energy complementarity by unifying the planning, coordinated control, and optimized operation of wind power, photovoltaic, and energy storage systems, achieving power balance and optimal resource allocation in both time and space dimensions. Ideally, wind-solar-storage coordinated dispatch should comprehensively consider meteorological forecast information, grid dispatch requirements, equipment operating status, and energy storage lifespan degradation characteristics. It should dynamically generate optimal output commands for each power generation unit through a combination of multi-timescale rolling optimization and event-driven coordination, maximizing new energy consumption and overall economic benefits while ensuring the safety and stability of the power system, and simultaneously considering the lifespan and health status of energy storage equipment.

[0004] However, existing wind, solar, and energy storage coordinated scheduling methods have significant shortcomings, making it difficult to meet the operational needs of large-scale new energy bases. On the one hand, wind power, photovoltaic, and energy storage systems are often managed independently or using simple timing rules, lacking a system-level spatiotemporal multi-energy complementary coordination mechanism. Power coordination between subsystems mainly relies on experience-based settings or fixed priority rules, failing to fully utilize the complementary characteristics of different energy resources for dynamic optimization. This results in limited overall efficiency of wind, solar, and energy storage joint operation, and frequent occurrences of wind and solar curtailment. On the other hand, although some technical solutions attempt to introduce centralized optimization scheduling strategies, their effectiveness highly depends on high-precision power prediction and complex large-scale mathematical optimization calculations. In actual operation, prediction errors are inevitable, and complex optimization algorithms are time-consuming to calculate, making it difficult to meet the needs of rapid power fluctuation smoothing and real-time frequency regulation response at the minute or even second level. The response lag problem is particularly prominent under sudden operating conditions such as extreme weather or grid failures. Furthermore, multi-energy coordination schemes based on fixed rules often struggle to simultaneously address multiple objectives, such as energy storage lifespan maintenance, state-of-charge balance control, and reactive power voltage support. Energy storage units age faster during frequent deep charge-discharge cycles, resulting in significant deterioration of their health and capacity decay. Simultaneously, grid connection voltage fluctuations are difficult to control effectively. These technical bottlenecks collectively lead to poor overall operational economy, shortened equipment lifespan, and insufficient grid-friendliness in existing renewable energy bases. Summary of the Invention

[0005] The present invention provides a multi-energy collaborative scheduling method for new energy bases based on wind, solar and energy storage coordination, in order to solve the technical problems of the existing centralized scheduling of wind, solar and energy storage, which relies heavily on high-precision power prediction and complex algorithms, resulting in lag response under sudden operating conditions, as well as the lack of complementary coordination between independent control of each system, leading to high curtailment rate, shortened lifespan of energy storage equipment and large grid voltage fluctuations.

[0006] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0007] This invention provides a multi-energy coordinated scheduling method for new energy bases based on wind, solar, and energy storage coordination. We construct a three-level coordinated architecture with hierarchical distribution: base layer, station layer, and equipment layer. This achieves the integration of multi-timescale rolling correction and event-triggered coordination. The specific method includes:

[0008] Step 1: Receive the wind and solar power prediction data and grid dispatch instructions from the base layer energy management platform located at the base layer, and generate the base value planning curves for each wind-solar-storage combined power generation unit and the voltage reference values ​​allocated to each station.

[0009] Step 2: By receiving ultra-short-term power prediction data, real-time energy storage state of charge data, and grid connection point voltage deviation data from the edge collaborative controller deployed at the power station layer, the total active power output command and the total reactive power output command of the power station are dynamically corrected; and scheduling decisions are executed based on the preset collaborative response logic triggered by the detected events.

[0010] Step 3: Through the energy storage converter, wind turbine inverter and photovoltaic inverter located in the equipment layer, rapid power sharing is achieved based on the high-speed communication network, and when the system is disturbed, the autonomous response mechanism is triggered by autonomously identifying frequency anomalies and voltage anomalies.

[0011] Furthermore, in step 1, the constraints for generating the base value planning curves of each wind-solar-storage combined power generation unit include the maximum power generation limit of wind power and photovoltaic power, the capacity limit and charging / discharging power limit of the energy storage system, the safe operating range of grid connection point voltage and frequency, and the power balance requirements of the grid dispatching command; the voltage reference value allocated to each station is determined by the base-level energy management platform based on the reactive voltage safety domain, and satisfies the safety constraints of the reactive voltage sensitivity matrix of the grid connection point voltage of each station determined based on the power flow calculation results of the entire network.

[0012] Furthermore, the multi-timescale rolling correction includes day-ahead scheduling for generating the base value plan curve at the daily level with a first preset time period, preset time-scale scheduling for rolling updates of the medium-term plan with a second preset time period, and minute-level rolling scheduling for dynamic adjustment of real-time output with a third preset time period; the third preset time period is shorter than the second preset time period, and the second preset time period is shorter than the first preset time period; the execution method of the dynamic correction of the total active power output command and the total reactive power output command of the power station is as follows: at the beginning of each rolling cycle, the power output plan of the power station is adjusted according to the latest ultra-short-term power prediction data, and when a power change is detected, the power output command is updated within a preset response time.

[0013] Furthermore, the detected events include power surge events, frequency anomaly events, voltage anomaly events, and communication failure events; when a power drop occurs at a certain power station, the triggering conditions for the edge collaborative controller to execute scheduling decisions include: the real-time power change rate of the power station exceeds a preset threshold, and the energy storage of adjacent power stations has available compensation capacity.

[0014] Furthermore, when determining the available compensation capacity, the calculation logic for the available charging power of the energy storage system currently constrained by the remaining energy and the rated power of the equipment is as follows: take the smaller value between the first charging calculation value and the second charging calculation value as the available charging power, wherein the first charging calculation value is the rated capacity of the energy storage system multiplied by the difference between the set upper limit value of the state of charge and the current state of charge, and then divided by the preset expected duration of power deficit compensation, and the second charging calculation value is the rated power of the energy storage system divided by the charging efficiency of the energy storage system;

[0015] The calculation logic for available discharge power is as follows: take the smaller value between the first discharge calculation value and the second discharge calculation value as the available discharge power, wherein the first discharge calculation value is the rated capacity of the energy storage system multiplied by the difference between the current state of charge and the set lower limit of the state of charge, and then divided by the preset expected duration of power deficit compensation, and the second discharge calculation value is the rated power of the energy storage system multiplied by the discharge efficiency of the energy storage system.

[0016] When the available energy storage capacity of this station is insufficient, the edge collaborative controller sends a support request message containing information such as the power deficit value, expected response time, and compensation duration to the adjacent stations.

[0017] Furthermore, when generating reference values ​​for energy storage charge and discharge power, the edge collaborative controller comprehensively considers the state of charge (SOC) and health decay function. The health decay function is used to generate a health correction coefficient. The calculation logic of the health correction coefficient is as follows: calculate the product of the cumulative equivalent charge and discharge depth with a first empirical coefficient, the product of the cumulative number of cycles with a second empirical coefficient, and the product of the average operating temperature with a third empirical coefficient; add a constant 1 to the sum of the three products to obtain the denominator; divide the constant 1 by the denominator to obtain the health correction coefficient; when the SOC of the energy storage unit is lower than the lower limit of the SOC, the edge collaborative controller generates a priority charging command; when the SOC of the energy storage unit is higher than the upper limit of the SOC, a priority discharging command is generated; when the SOC of the energy storage unit is between the lower limit of the SOC and the upper limit of the SOC, the charge and discharge priority is adjusted according to the health correction coefficient.

[0018] Furthermore, in step 3, the autonomous response mechanism includes an inertia support response and a primary frequency regulation response. The inertia support response is as follows: in response to rapid changes in system frequency, the energy storage converter, the wind turbine inverter, and the photovoltaic inverter automatically inject an active power increment proportional to the frequency change rate based on their own rotational inertia or virtual inertia characteristics. The primary frequency regulation response is as follows: in response to steady-state deviations in system frequency, when the system frequency deviation exceeds a preset frequency dead zone, the energy storage converter, the wind turbine inverter, and the photovoltaic inverter automatically adjust their active power output to participate in primary frequency regulation according to a frequency regulation coefficient determined based on the equipment's rated capacity and operating status.

[0019] Furthermore, the method also includes a step of achieving reactive power and voltage coordination. The specific execution logic is as follows: the edge coordination controller collects the real-time measured value of the grid connection point voltage, calculates the deviation from the voltage reference value allocated to the power station, and queries the remaining reactive power capacity of each wind turbine inverter, photovoltaic inverter, and energy storage converter in the power station; the edge coordination controller allocates reactive power increments to each device in descending order of reactive power and voltage sensitivity, and when the reactive power capacity of a high-sensitivity device reaches its upper limit, it allocates the reactive power increments to low-sensitivity devices in sequence.

[0020] Furthermore, when achieving reactive voltage coordination, if the deviation is within a preset deviation threshold range, a local rapid control strategy is adopted; if the deviation exceeds the deviation threshold range, a base-level reactive power reserve coordination mechanism is triggered. The execution process of the base-level reactive power reserve coordination mechanism is as follows: when the edge coordination controller detects that the voltage deviation exceeds a preset safety limit, it sends a reactive power support request to the base-level energy management platform, and the base-level energy management platform allocates reactive power support power from adjacent power stations according to the global optimization results.

[0021] Furthermore, the base layer and the station layer use a fiber optic Ethernet network as the backbone communication network to support the issuance of dispatch instructions and plans as well as the uploading of real-time status data; the station layer and the equipment layer use a communication protocol based on the IEC 61850 standard, and adopt a substation event GOOSE message mechanism oriented towards general objects to realize instruction transmission and event reporting; the converters and inverters within the equipment layer use point-to-point high-speed communication or a star topology based on switches for communication.

[0022] Compared with the prior art, the beneficial effects of the present invention are:

[0023] (1) By establishing a three-level collaborative architecture of base layer, station layer and equipment layer, multi-time scale rolling correction and event-triggered collaboration are adopted to replace single centralized large-scale optimization, reducing the dependence on high-precision power prediction and complex calculation; under the condition of prediction error and sudden working conditions, each subsystem works closely together to quickly compensate for power deficit, solving the problem of response lag under extreme weather or grid failure, significantly improving the robustness and stability of the system, and effectively reducing the curtailment rate.

[0024] (2) In the scheduling decision of edge collaborative control, the generation of energy storage charging and discharging power reference value fully integrates the state of charge and the health decay function based on multi-parameter construction, so that energy storage units with low state of charge and good health are given priority to charging and those with high state of charge are given priority to discharging, avoiding deep overcharging and over-discharging, and balancing the losses of each energy storage unit throughout the entire life cycle.

[0025] (3) Through the autonomous response mechanism of the underlying equipment, the energy storage converter and inverter can autonomously identify frequency anomalies and inject active power increments instantaneously to complete inertia support and primary frequency regulation without waiting for upper-level instructions, thus achieving millisecond-level sharing response.

[0026] (4) Through the reactive voltage coordination mechanism, the station controller integrates the voltage deviation at the grid connection point and uses the remaining reactive capacity of the inverter and energy storage converter to distribute the reactive increment according to the sensitivity order. This achieves close coordination between local rapid voltage control and base-level reactive power reserve, which not only controls the voltage fluctuation at the grid connection point but also avoids the increase in losses caused by remote reactive power transmission.

[0027] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, embodiments of the present invention are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a three-level collaborative architecture diagram of the multi-energy collaborative scheduling method for new energy bases based on wind, solar and energy storage coordination in the embodiment.

[0030] Figure 2 This is a schematic diagram of the scheduling principle framework for multi-timescale rolling correction and event triggering coordination in the embodiment.

[0031] Figure 3This is a flowchart of the collaborative response logic of the edge collaborative controller at the station layer in the embodiment. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0033] Example 1

[0034] This invention provides a multi-energy coordinated scheduling method for new energy bases based on wind, solar, and energy storage coordination, such as... Figure 1 As shown, it constructs a hierarchical, distributed three-level collaborative architecture of base-station-equipment, achieving an organic integration of multi-timescale rolling correction and event-triggered collaboration. The entire scheduling system consists of a base-level energy management platform, a station-level edge collaborative controller, and an equipment-level intelligent energy storage converter. Data interaction and command transmission between the three levels are achieved through a high-speed communication network, forming a closed-loop feedback collaborative control system. The specific method of this invention is as follows:

[0035] Step 1: Establish a base-level energy management platform.

[0036] As the top-level decision-making unit of the entire dispatch system, the energy management platform undertakes the core functions of global optimization and coordinated dispatch. The platform uses a preset time period as the benchmark for power forecasting and receiving dispatch instructions, ranging from 8 to 24 hours (a timescale from several hours to the day-ahead). The platform continuously receives network-wide wind and solar power forecast data from the meteorological forecasting system, including wind speed forecast curves, solar irradiance forecast curves, and installed capacity information for each wind farm and photovoltaic power station for the next 24 to 72 hours. Simultaneously, the platform receives grid dispatch instructions from the grid dispatch center, which include planned active power, planned reactive power, and voltage control targets for each time period.

[0037] In this embodiment, the energy management platform generates baseline planning curves for each wind-solar-storage combined power generation unit based on received network-wide wind and solar power forecast data and grid dispatch instructions, using a multi-objective optimization algorithm. The baseline planning curves include target sequences for active power output, reactive power output, and the planned charging and discharging power of the energy storage system for each power station. The generation process of the baseline planning curves comprehensively considers the following constraints: maximum renewable power limits for wind and solar power, capacity and charging / discharging power limits for the energy storage system, safe operating range of grid connection voltage and frequency, and power balance requirements of the grid dispatch instructions.

[0038] In this embodiment, the energy management platform simultaneously allocates voltage reference values ​​to each power station based on the reactive power-voltage safety domain. The reactive power-voltage safety domain is defined as the allowable fluctuation range of the grid-connected voltage under the current system operating conditions to ensure voltage stability. Based on the power flow calculation results of the entire network, the energy management platform determines the reactive power-voltage sensitivity matrix of the grid-connected voltage for each power station, and then allocates a voltage reference value that meets safety constraints to each power station. When performing quasi-steady-state coordination optimization, the platform comprehensively weighs factors such as the accuracy of the overall network wind and solar power forecast, the degree of fulfillment of grid dispatch instructions, the thermal stability of equipment operation, and the economics of energy storage lifespan degradation, and obtains the globally optimal or suboptimal dispatch scheme through a weighted multi-objective optimization method.

[0039] Step 2: Deploy the edge collaboration controller at the site layer.

[0040] As a real-time scheduling and decision-making unit at the site level, the edge collaborative controller operates in a preset time granularity rolling mode, with the preset time granularity ranging from 1 minute to 15 minutes, i.e., a minute-level rolling mode. The controller continuously receives three types of real-time information streams: ultra-short-term power prediction data, real-time energy storage state-of-charge data, and grid connection point voltage deviation data.

[0041] In this embodiment, the prediction duration for ultra-short-term power forecast data is 15 minutes to 4 hours, the update cycle is 1 minute to 5 minutes, and the prediction accuracy is required to reach over 90%. Real-time state of charge (SBC) data for energy storage reflects the current remaining percentage of energy in each energy storage unit, typically presented as a value from 0 to 100%, with an update frequency of no less than 1 second. Grid connection point voltage deviation data is the difference between the actual measured voltage value and the rated voltage value, used to assess the degree of voltage deviation.

[0042] In this embodiment, the edge collaborative controller dynamically corrects the total active power output command and the total reactive power output command of the power station based on the three types of real-time data mentioned above. The controller adopts a rolling optimization strategy, adjusting the power station's output plan according to the latest ultra-short-term power forecast data at the beginning of each rolling cycle; when a power change is detected, the controller completes the update of the output command within a preset response time, with an update delay of no more than 10 seconds.

[0043] In this embodiment, when a power station experiences a sharp drop in power due to changes in environmental factors, the edge collaborative controller executes scheduling decisions through predefined collaborative response logic. Changes in environmental factors include a sharp drop in photovoltaic output due to cloud cover and fluctuations in wind power output due to sudden changes in wind conditions. For example... Figure 3As shown, the triggering conditions for the collaborative response logic include two parallel conditions: the real-time power change rate of the power station exceeds a preset threshold, and the energy storage of adjacent power stations has available compensation capacity. The default value of the preset threshold is a change rate of 15% to 25% of the installed capacity per minute, which is adaptively adjusted based on historical operating data and power station characteristics. The historical operating data analysis period is the operating records of the past 30 days, and the power station characteristics include parameters such as the power station's installed capacity, equipment response speed, and communication latency.

[0044] In this embodiment, the edge collaboration controller employs a hierarchical invocation strategy when executing collaborative responses. For example... Figure 3 As shown, the controller first determines whether the available compensation power of the energy storage system meets the power deficit compensation requirements. To ensure the safety of the energy storage system and constrained by the physical boundaries of the equipment, its available charging and discharging power must be constrained by both the current remaining energy and the rated power of the equipment. Available charging power and available discharge power The calculation formula is as follows:

[0045]

[0046]

[0047] In the formula, This refers to the rated capacity of the energy storage system. This refers to the rated power of the energy storage system. The current state of charge; and These are the set upper and lower limits of the state of charge; and These refer to the charging and discharging efficiencies of the energy storage system, respectively. The preset expected duration of power deficit compensation (such as the time granularity of the rolling scheduling cycle).

[0048] When the available energy storage capacity at this station is insufficient, the controller sends a support request message to adjacent stations. The message includes information such as the power deficit value, the expected response time, and the duration of compensation. Upon receiving the support request, the adjacent stations decide whether to respond to the support request based on their own energy storage state of charge and planned output.

[0049] In this embodiment, the generation process of the energy storage charge / discharge power reference value comprehensively considers two core factors: state of charge and health decay function. The health decay function calculates a correction coefficient based on parameters such as the cumulative depth of charge / discharge, number of cycles, and temperature of the energy storage unit. The cumulative depth of charge / discharge is measured in ampere-hours and obtained through integration; the number of cycles is the cumulative number of times the energy storage unit completes one full charge / discharge cycle; and the temperature parameter is the real-time measured temperature value from the energy storage battery management system. The correction coefficient is calculated using the following function form:

[0050]

[0051] in, This is a health correction factor. To accumulate equivalent depth of charge and discharge, To accumulate the number of loops, The average operating temperature, , , This is an empirical coefficient. When the state of charge (SBC) of an energy storage unit is below the lower limit of the SBC value, the controller generates a priority charging command; when the SBC value is above the upper limit of the SBC value, the controller generates a priority discharging command; when the SBC value is between the upper and lower limits, the controller adjusts the charging and discharging priorities according to the health correction coefficient to ensure that each energy storage unit achieves balanced loss throughout its entire life cycle.

[0052] Step 3: Implement fast power sharing at the device layer.

[0053] Energy storage converters, wind turbine inverters, and photovoltaic inverters constitute the core execution units of the equipment layer, achieving rapid power sharing through a high-speed communication network. The high-speed communication network employs the GOOSE message mechanism based on the IEC 61850 standard, with message transmission latency controlled within 4ms and communication reliability exceeding 99.99%. The high-speed communication network supports both point-to-point and multicast communication modes. Point-to-point mode is used for reliable transmission of important commands, while multicast mode is used for broadcasting and distributing batch status information.

[0054] In this embodiment, the device-level fast power sharing mechanism achieves millisecond-level coordinated allocation of active and reactive power. When the system experiences disturbances or requires islanding operation, the underlying devices trigger a response by autonomously identifying frequency and voltage anomalies. The threshold for determining frequency anomalies is a system frequency deviating from the 50Hz rated frequency by more than 0.03Hz to 0.05Hz; the threshold for determining voltage anomalies is a grid connection point voltage deviating from the rated voltage by more than 3% to 5% of the rated value.

[0055] In this embodiment, the underlying device autonomous response mechanism includes two levels: inertia support response and primary frequency regulation response. The inertia support response addresses rapid changes in system frequency. Based on their own rotational inertia or virtual inertia characteristics, the energy storage converter and inverter automatically inject active power increments proportional to the rate of frequency change, with a response delay not exceeding 20ms. The primary frequency regulation response addresses steady-state deviations in system frequency. When the system frequency deviation exceeds a preset frequency dead zone, the energy storage converter and inverter automatically adjust their active power output to participate in primary frequency regulation according to a preset frequency regulation coefficient. The preset frequency dead zone is in the range of 0.02Hz to 0.033Hz, and the frequency regulation coefficient is determined based on the equipment's rated capacity and operating status. The voltage regulation response mechanism is similar to the primary frequency regulation response. When the grid connection point voltage deviation exceeds a preset voltage dead zone, the energy storage converter and inverter automatically adjust their reactive power output to participate in voltage regulation. The preset voltage dead zone is in the range of 1% to 2% of the rated voltage.

[0056] Step 4: Achieve reactive power and voltage coordination.

[0057] As the main body responsible for reactive power and voltage coordination, the station controller comprehensively processes the grid connection point voltage deviation signal, the reactive power capacity information of each inverter and energy storage converter within the station, and the base-level reactive power reserve strategy. The station controller first collects the real-time measured value of the grid connection point voltage and calculates the deviation from the target voltage reference value; then it queries the remaining reactive power capacity of each inverter and energy storage converter, which is equal to the rated reactive power capacity of the equipment minus the current actual reactive power output; finally, it allocates the reactive power increment according to the reactive power and voltage sensitivity order.

[0058] In this embodiment, reactive power voltage coordination allocates reactive power increments sequentially from high to low reactive power voltage sensitivity. Devices with higher sensitivity take priority in reactive power regulation tasks, and the sensitivity matrix is ​​obtained through offline power flow calculation and stored in the station controller. When the reactive power capacity of a high-sensitivity device reaches its upper limit, the controller sequentially allocates reactive power increments to lower-sensitivity devices. The voltage deviation threshold is a preset percentage range of the rated voltage, typically set to 3% to 5% of the rated voltage. When the voltage deviation is within the threshold range, a local fast control strategy is employed; when the voltage deviation exceeds the threshold range, a base-level reactive power reserve coordination mechanism is triggered.

[0059] In this embodiment, the base-level reactive power reserve coordination mechanism is uniformly coordinated by the energy management platform. Based on the overall network reactive power and voltage distribution, the platform reserves a certain proportion of reactive power reserve capacity to cope with severe voltage fluctuations. The allocation of reactive power reserve capacity follows the principle of economy, minimizing reactive power transmission losses while meeting voltage safety constraints. When the station controller detects a voltage deviation exceeding a preset safety limit, it sends a reactive power support request to the energy management platform. The platform allocates reactive power support power from adjacent stations based on global optimization results, achieving effective coordination between local rapid voltage control and base-level reactive power reserve.

[0060] In this embodiment, as Figure 2 As shown, the entire scheduling method operates through a combination of multi-timescale rolling correction and event-triggered coordination. Multi-timescale rolling correction includes three timescales: day-ahead scheduling, preset timescale scheduling, and minute-level rolling scheduling. Day-ahead scheduling has a planning period of 24 hours and a time resolution of 15 minutes to 1 hour, used to generate daily baseline planning curves; preset timescale scheduling has a planning period of 4 to 8 hours and a time resolution of 5 to 15 minutes, used for rolling updates of medium-term plans; minute-level rolling scheduling has a planning period of 15 minutes to 1 hour and a time resolution of 1 minute, used for dynamic adjustment of real-time output. Event-triggered coordination includes four types of trigger sources: power surge events, frequency anomaly events, voltage anomaly events, and communication failure events. The trigger condition for a power surge event is that the real-time power change rate exceeds a set threshold; the trigger condition for a frequency anomaly event is that the system frequency deviation exceeds the frequency dead zone; the trigger condition for a voltage anomaly event is that the voltage deviation at the grid connection point exceeds the voltage dead zone; and the trigger condition for a communication failure event is that the communication link between the upper and lower levels is interrupted or the communication quality deteriorates.

[0061] In this embodiment, the three-tiered collaborative mechanism forms a closed-loop feedback loop through data interaction and command transmission. The base-level energy management platform issues baseline planning curves and voltage reference values ​​to the field station level. The field station level edge collaborative controller feeds back the real-time operating status to the base level, and the base level makes rolling adjustments to the plan based on the feedback information. The field station level issues output commands to the equipment level, and the equipment level reports power sharing status and abnormal events to the field station level, which adjusts the collaborative control strategy accordingly. This hierarchical collaborative architecture reduces the reliance on centralized large-scale optimization, significantly improving local response speed and system robustness while ensuring the overall optimization effect of the system.

[0062] Example 2

[0063] This embodiment provides a specific application scenario for a multi-energy coordinated scheduling method for new energy bases based on wind, solar, and energy storage coordination. A detailed explanation is provided using a 500 MW integrated wind, solar, and energy storage base as an example. This base comprises three stations: wind farm A, photovoltaic power station B, and energy storage station C, with installed capacities of 200 MW, 200 MW, and 100 MW / 200 MWh, respectively.

[0064] In this embodiment, the execution process of step 1 is as follows: The energy management platform receives wind speed and solar irradiance forecast data for the next 72 hours from the meteorological forecast system at 0:00 AM every day. Assuming that the daily forecast shows that photovoltaic output will be at its peak between 10:00 AM and 2:00 PM, with a projected maximum photovoltaic output of 180 MW; and that wind power output will be at its peak between 6:00 PM and 10:00 PM, with a projected maximum wind power output of 160 MW. The energy management platform, combining the planned active power values ​​for each time period specified in the grid dispatch instructions, and comprehensively considering the capacity constraints and operating costs of energy storage station C, generates the baseline planned curves for each station for the day. Specifically, from 10:00 AM to 2:00 PM, the platform sets the active power output target for photovoltaic power station B at 150 MW and the charging power target for energy storage power station C at 30 MW to absorb excess photovoltaic power; from 6:00 PM to 10:00 PM, the platform sets the active power output target for wind power station A at 140 MW and the discharge power target for energy storage power station C at 20 MW to support the evening peak power supply demand.

[0065] In this embodiment, step 2 is executed as follows: The edge collaborative controller at the power station level operates continuously with a rolling cycle of 5 minutes. Assuming that at 1 PM, the controller detects that the real-time output of photovoltaic power station B drops sharply from 150 MW to 80 MW, with a power change rate of 35 MW / min, exceeding the preset threshold of 20 MW / min, the controller immediately triggers the collaborative response logic. First, it determines whether the available discharge capacity of the local energy storage at photovoltaic power station B meets the deficit compensation requirements. Assuming that photovoltaic power station B has a supporting energy storage capacity of 30 MW / 60 MWh and a current state of charge of 60%, the available discharge capacity is 18 MWh, corresponding to a discharge power of approximately 30 MW. The power deficit is 70 MW, indicating insufficient local energy storage capacity. The controller sends a support request to the adjacent energy storage station C. Upon receiving the support request, energy storage station C checks its own operating status. The current state of charge is 70%, and the available discharge capacity is approximately 45 MWh, corresponding to approximately 50 MW of discharge power. It responds to the support request and outputs 30 MW of discharge power. Wind farm A simultaneously increased its output by 20 MW to compensate for a 70 MW power deficit. Throughout the coordinated response process, the power compensation delay did not exceed 30 seconds, effectively mitigating minute-level power fluctuations.

[0066] In this embodiment, the generation process of the energy storage charge / discharge power reference value comprehensively considers the state of charge and the health decay function. The controller calculates the health correction coefficient based on the cumulative charge / discharge depth, cycle count, and temperature parameters of each energy storage unit. When the cumulative equivalent charge / discharge depth of a certain energy storage unit reaches more than 80% of the design limit, the controller reduces the weight of its charge / discharge power reference value and prioritizes the output of energy storage units with higher health, thereby achieving a balance of losses among the energy storage units throughout their entire life cycle.

[0067] In this embodiment, step 3 is executed as follows: The energy storage converter and inverter at the equipment layer communicate at the millisecond level via GOOSE messages based on IEC61850. Assume the system frequency suddenly drops to 49.92Hz, deviating from the rated frequency by more than 0.05Hz (dead zone threshold). The energy storage converter at energy storage station C responds immediately, autonomously identifying the frequency anomaly and injecting an increase in active power according to the preset primary frequency regulation characteristics. The frequency regulation power is 4% to 5% of the rated capacity, with a response delay of no more than 50ms. The inverter at photovoltaic station B synchronously participates in the primary frequency regulation, injecting an increase in active power of approximately 2% of the rated capacity. The entire primary frequency regulation process does not require waiting for upper-level commands and is entirely completed autonomously by the underlying equipment, effectively supporting the rapid recovery of the system frequency.

[0068] In this embodiment, step 4 is executed as follows: The station controller monitors the grid connection point voltage in real time, assuming the voltage drops to 96% of the rated voltage, exceeding the preset 3% deviation threshold. The controller queries the remaining reactive power capacity of each inverter and energy storage converter, and calculates the reactive power voltage sensitivity matrix. Assuming the photovoltaic inverter has the highest reactive power voltage sensitivity, its remaining reactive power capacity is prioritized. The photovoltaic inverter currently has an active power output of 80 MW, a rated reactive power capacity of 32 MVA, an output reactive power of 16 MVA, and a remaining reactive power capacity of 16 MVA. The controller first increases the reactive power output of the photovoltaic inverter by 12 MVA. If the voltage still does not recover, the remaining reactive power capacity of the energy storage converter is then used. Through this strategy of allocating reactive power increments according to sensitivity order, local rapid voltage control is achieved, avoiding the increased losses caused by remote reactive power transmission.

[0069] Example 3

[0070] This embodiment provides a specific implementation of the communication architecture and anomaly handling mechanism for a multi-energy collaborative scheduling method for new energy bases based on wind, solar and energy storage coordination.

[0071] In this embodiment, the three-tier collaborative architecture's communication network adopts a layered design. The base station layer and the field station layer use a fiber optic Ethernet network as the backbone communication network, with a communication bandwidth of no less than 1Gbps and a communication latency of no more than 100ms, supporting the issuance of day-ahead scheduling instructions and medium-term rolling plans, as well as the uploading of real-time status data. The field station layer and the equipment layer use Fast Ethernet communication based on the IEC 61850 standard, with a communication bandwidth of 100Mbps, and employ the GOOSE message mechanism to achieve millisecond-level instruction transmission and event reporting. Within the equipment layer, each converter communicates via point-to-point high-speed communication or a switch-based star topology, with communication latency controlled within 4ms.

[0072] In this embodiment, the event handling process of the event-triggered coordination mechanism is designed as follows: When the edge coordination controller detects a power surge event, the controller immediately initiates a preset coordination response process, completing deficit calculation, capacity determination, and support request transmission within 50ms. When a neighboring station receives a support request, it completes response decision and power output adjustment within 100ms. When the controller detects a frequency anomaly event, the energy storage converter and inverter complete an inertia support response within 20ms and a frequency regulation response within 100ms. When the controller detects a voltage anomaly event, the inverter and energy storage converter complete reactive power output adjustment within 50ms. When the controller detects a communication failure event, it immediately switches to local autonomous operation mode, maintaining the basic operating functions of the equipment based on local measurement data and preset strategies, while simultaneously sending fault alarm information to the upper layer.

[0073] In this embodiment, the sensitivity calculation and allocation strategy for reactive power voltage coordination is specifically implemented as follows: the sensitivity matrix is ​​obtained through offline power flow calculation, taking into account the current operating point of the system and the network topology. Elements of the sensitivity matrix... Represents a node The impact of reactive power changes on nodes The impact of voltage. During reactive power allocation, the controller first normalizes the sensitivity matrix column-by-column to obtain normalized sensitivity coefficients; then, it sequentially allocates the reactive power capacity of each device according to the normalized sensitivity coefficients from largest to smallest. Within each allocation cycle, the controller updates the sensitivity matrix every 5 minutes to adapt to changes in the system operating point.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-energy coordinated scheduling method for new energy bases based on wind, solar, and energy storage coordination, characterized in that, A three-tiered collaborative architecture consisting of a base layer, a field station layer, and an equipment layer is constructed to achieve the integration of multi-timescale rolling correction and event-triggered collaboration. The method includes: Step 1: Receive the wind and solar power prediction data and grid dispatch instructions from the base layer energy management platform located at the base layer, and generate the base value planning curves for each wind-solar-storage combined power generation unit and the voltage reference values ​​allocated to each station. Step 2: By receiving ultra-short-term power prediction data, real-time energy storage state of charge data, and grid connection point voltage deviation data from the edge collaborative controller deployed at the power station layer, the total active power output command and the total reactive power output command of the power station are dynamically corrected; and scheduling decisions are executed based on the preset collaborative response logic triggered by the detected events. Step 3: Through the energy storage converter, wind turbine inverter and photovoltaic inverter located in the equipment layer, rapid power sharing is achieved based on the high-speed communication network, and when the system is disturbed, the autonomous response mechanism is triggered by autonomously identifying frequency anomalies and voltage anomalies.

2. The method according to claim 1, characterized in that, In step 1, the constraints for generating the base value planning curves for each wind-solar-storage combined power generation unit include the maximum power generation limit for wind and solar power, the capacity limit and charging / discharging power limit for the energy storage system, the safe operating range of grid connection point voltage and frequency, and the power balance requirements of the grid dispatching instructions. The voltage reference value allocated to each power station is determined by the base-level energy management platform based on the reactive voltage safety domain, and satisfies the safety constraints of the reactive voltage sensitivity matrix of the grid connection point voltage of each power station determined based on the power flow calculation results of the entire network.

3. The method according to claim 1, characterized in that, The multi-timescale rolling correction includes day-ahead scheduling that generates the base value plan curve at the daily level with a first preset time period, preset time-scale scheduling that performs rolling updates of the medium-term plan with a second preset time period, and minute-level rolling scheduling that performs dynamic adjustments to real-time power output with a third preset time period; the third preset time period is shorter than the second preset time period, and the second preset time period is shorter than the first preset time period; the execution method of the dynamic correction of the total active power output command and the total reactive power output command of the power station is as follows: at the beginning of each rolling cycle, the power output plan of the power station is adjusted according to the latest ultra-short-term power prediction data, and when a power change is detected, the power output command is updated within a preset response time.

4. The method according to claim 1, characterized in that, The detected events include power surge events, frequency anomaly events, voltage anomaly events, and communication failure events. When a power drop occurs at a certain power station, the triggering conditions for the edge collaborative controller to execute scheduling decisions include: the real-time power change rate of the power station exceeds a preset threshold, and the energy storage of adjacent power stations has available compensation capacity.

5. The method according to claim 4, characterized in that, When determining the available compensation capacity, the calculation logic for the available charging power of the energy storage system, which is currently constrained by the remaining energy and the rated power of the equipment, is as follows: take the smaller value between the first charging calculation value and the second charging calculation value as the available charging power. The first charging calculation value is the rated capacity of the energy storage system multiplied by the difference between the set upper limit of the state of charge and the current state of charge, and then divided by the preset expected duration of power deficit compensation. The second charging calculation value is the rated power of the energy storage system divided by the charging efficiency of the energy storage system. The calculation logic for available discharge power is as follows: take the smaller value between the first discharge calculation value and the second discharge calculation value as the available discharge power, wherein the first discharge calculation value is the rated capacity of the energy storage system multiplied by the difference between the current state of charge and the set lower limit of the state of charge, and then divided by the preset expected duration of power deficit compensation, and the second discharge calculation value is the rated power of the energy storage system multiplied by the discharge efficiency of the energy storage system. When the available energy storage capacity of this station is insufficient, the edge collaborative controller sends a support request message containing information such as the power deficit value, expected response time, and compensation duration to the adjacent stations.

6. The method according to claim 1, characterized in that, When generating reference values ​​for energy storage charging and discharging power, the edge collaborative controller comprehensively considers the state of charge and the health decay function. The health decay function is used to generate a health correction coefficient. The calculation logic of the health correction coefficient is as follows: calculate the product of the cumulative equivalent charging and discharging depth and the first empirical coefficient, the product of the cumulative number of cycles and the second empirical coefficient, and the product of the average operating temperature and the third empirical coefficient. Add a constant 1 to the sum of the three products to obtain the denominator; divide the constant 1 by the denominator to obtain the health correction coefficient; When the state of charge of the energy storage unit is lower than the lower limit of the state of charge, the edge collaborative controller generates a priority charging command; when the state of charge of the energy storage unit is higher than the upper limit of the state of charge, it generates a priority discharging command; when the state of charge of the energy storage unit is between the lower limit of the state of charge and the upper limit of the state of charge, the charging and discharging priority is adjusted according to the health correction coefficient.

7. The method according to claim 1, characterized in that, In step 3, the autonomous response mechanism includes inertia support response and primary frequency modulation response; The inertia support response is as follows: in response to rapid changes in system frequency, the energy storage converter, the wind turbine inverter, and the photovoltaic inverter automatically inject an active power increment proportional to the frequency change rate based on their own rotational inertia or virtual inertia characteristics. The primary frequency regulation response is as follows: In response to the steady-state deviation of the system frequency, when the system frequency deviation exceeds the preset frequency dead zone, the energy storage converter, the wind turbine inverter, and the photovoltaic inverter automatically adjust their active power output to participate in the primary frequency regulation according to the frequency regulation coefficient determined based on the rated capacity and operating status of the equipment.

8. The method according to claim 1, characterized in that, The method further includes a step of achieving reactive power and voltage coordination. The specific execution logic is as follows: the edge coordination controller collects the real-time measured value of the grid connection point voltage, calculates the deviation from the voltage reference value allocated to the station, and queries the remaining reactive power capacity of each wind turbine inverter, photovoltaic inverter, and energy storage converter in the station; the edge coordination controller allocates reactive power increments to each device in descending order of reactive power and voltage sensitivity, and when the reactive power capacity of a high-sensitivity device reaches its upper limit, it allocates the reactive power increments to low-sensitivity devices in turn.

9. The method according to claim 8, characterized in that, When achieving reactive power voltage coordination, if the deviation is within a preset deviation threshold range, a local rapid control strategy is adopted; if the deviation exceeds the deviation threshold range, a base-level reactive power reserve coordination mechanism is triggered. The execution process of the base-level reactive power reserve coordination mechanism is as follows: when the edge coordination controller detects that the voltage deviation exceeds the preset safety limit, it sends a reactive power support request to the base-level energy management platform. The base-level energy management platform allocates reactive power support power to adjacent power stations according to the global optimization results.

10. The method according to claim 1, characterized in that, The base layer and the station layer use a fiber optic Ethernet network as the backbone communication network to support the issuance of scheduling instructions and plans as well as the uploading of real-time status data. The communication protocol between the station layer and the equipment layer is based on the IEC 61850 standard, and the GOOSE message mechanism for substation events, which is oriented towards general objects, is used to realize command transmission and event reporting. The converters and inverters within the equipment layer use point-to-point high-speed communication or star networking based on switches.

Images