A virtual power plant distributed energy resource optimal scheduling method and system

Abstract

The application discloses a kind of virtual power plant distributed energy resource optimization scheduling method and system, belong to virtual power plant technical field.The method defines multiple combination abnormal state in advance and preconfigures corresponding scheduling strategy, real-time acquisition operating state data determines current feature, determines target abnormal state by comparing with pre-defined state, executes corresponding strategy and sends control instruction to standby equipment.The application also proposes abnormal matching based on importance coefficient, strategy correction based on equipment availability, safety screening based on production influence verification and coordinated control mechanism based on power response timing, solves the problem of multiple concurrent abnormal response delay, realizes fast and accurate scheduling, and improves power grid stability.

Description

Technical Field

[0001] This application relates to the field of virtual power plant technology, and more specifically, to a method and system for optimizing the scheduling of distributed energy resources in a virtual power plant. Background Technology

[0002] In modern power systems, virtual power plants serve as an innovative energy management model. In practice, virtual power plants face the challenge of efficiently aggregating distributed energy resources and rapidly responding to grid demands. Traditional energy management systems struggle to flexibly coordinate unstable energy sources such as photovoltaic power generation equipment, potentially leading to grid instability and increased energy trading costs.

[0003] Specifically, when a virtual power plant encounters multiple concurrent unexpected events, such as unforeseen extreme weather causing a sudden failure in photovoltaic power generation, while critical controllable load resources (such as energy storage devices or production line equipment) abruptly and without warning exit the response due to their own commercial reasons, the existing forecasting and planning-based scheduling model will fail. After detecting a significant deviation between power generation and planned values, the system typically activates an automatic compensation mechanism, attempting to cope by recalculating the global optimization.

[0004] However, this approach of recalculating for optimization presents significant technical challenges. It requires communicating with numerous dispersed and heterogeneous secondary resources (such as auxiliary equipment) to obtain their real-time operating status and adjustable potential. Since each subsystem has different communication protocols and response speeds, information aggregation itself is time-consuming. The feedback information may also carry various constraints, requiring the control system to perform a complex recalculation upon receiving this fragmented information in order to piece together a feasible combination from the remaining resources to fill the power gap. The entire process of problem detection, re-querying, recalculation, and issuing new instructions is excessively time-consuming, leading to response delays and an inability to effectively fill the power gap within the grid's required timeframe. Ultimately, this results in scheduling task failure, economic penalties, and damage to its reputation as a reliable regulating resource for the grid.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application discloses a method and system for optimizing the scheduling of distributed energy resources in a virtual power plant. It aims to solve the problem that when existing virtual power plants face multiple concurrent abnormal states, the traditional scheduling mode based on prediction and planning is slow to respond, inefficient, and unable to make up for the power gap in a timely and effective manner, resulting in the failure of scheduling tasks.

[0007] The technical solution of this application is as follows: In a first aspect, this application discloses a method for optimizing the scheduling of distributed energy resources in a virtual power plant, comprising the following steps: predefining multiple combined abnormal states, each combined abnormal state corresponding to a concurrent state in which multiple devices in the distributed energy resources simultaneously malfunction; preconfiguring a corresponding scheduling strategy for each combined abnormal state, wherein the scheduling strategy pre-specifies the standby devices to be called and the power regulation parameters of each standby device; the distributed energy resources include photovoltaic power generation equipment, energy storage equipment, production line equipment, and auxiliary power equipment; real-time acquisition of the operating status data of the distributed energy resources, and determining the current operating status characteristics based on the operating status data; comparing the current operating status characteristics with the predefined combined abnormal states to determine the target combined abnormal state corresponding to the currently occurring concurrent state; executing the scheduling strategy corresponding to the target combined abnormal state, and sending control commands to the corresponding standby devices according to the standby devices and power regulation parameters pre-specified in the scheduling strategy.

[0008] Furthermore, the combined abnormal states include: a first combined abnormal state: the output power of the photovoltaic power generation equipment decreases at a rate exceeding a first preset threshold, and the charging and discharging control connection of the energy storage equipment is interrupted; a second combined abnormal state: the output power of the photovoltaic power generation equipment decreases at a rate exceeding the first preset threshold, and the increase in the power load of the production line equipment exceeds the second preset threshold.

[0009] Based on this, the current operating status characteristics are determined according to the operating status data, including: monitoring the output power decline rate of the photovoltaic power generation equipment, and generating a power generation decline characteristic when the output power decline rate exceeds the first preset threshold; monitoring the charging and discharging control connection status of the energy storage equipment, and generating an energy storage equipment disconnection characteristic when the charging and discharging control connection is interrupted; and monitoring the change in the power load of the production line equipment, and generating a load increase characteristic when the increase in power load exceeds the second preset threshold.

[0010] In some preferred embodiments, the scheduling strategy pre-specifies the backup equipment to be called up and the power adjustment parameters of each backup equipment, including: determining the type of backup equipment to be called up, wherein the backup equipment is the auxiliary power equipment, including at least one of a central air conditioning system, cooling pump, lighting system or non-core production line; determining the power reduction or power increase of each backup equipment; and determining the priority order for sending control commands to each backup equipment.

[0011] Furthermore, the current operating state features are compared with predefined combined abnormal states to determine the target combined abnormal state corresponding to the current concurrent state. This includes: calculating the similarity value between the current operating state features and the corresponding features in the combined abnormal state; configuring an importance coefficient for each feature in the combined abnormal state; weighting and summing the similarity values ​​according to the importance coefficient to obtain a matching value; sorting the matching values ​​of each combined abnormal state; and determining the combined abnormal state with the highest matching value that exceeds a preset threshold as the target combined abnormal state.

[0012] As a technical improvement, before executing the scheduling strategy corresponding to the abnormal state of the target combination, the following steps are also included: querying the enterprise management system for equipment availability information through the application programming interface; calculating the actual available adjustable power of each standby device based on the equipment availability information, and correcting the standby devices and their power adjustment parameters specified in the scheduling strategy accordingly.

[0013] Based on this, the revised scheduling strategy further includes: sending a simulation adjustment command to the production impact verification module deployed at each backup device, the simulation adjustment command containing the simulation adjustment range; receiving the safety verification result returned by the production impact verification module, the safety verification result including the safe adjustable power range and safety impact level of the backup device in the current production state, the safety impact level being determined by querying the coupling relationship between the backup device and the core production process; and removing backup devices with a safe adjustable power range of zero or a safety impact level exceeding a preset level from the scope of the scheduling strategy.

[0014] To improve the scheme, the scheduling strategy also includes power response timing parameters for each backup device. These power response timing parameters include the startup delay time, power adjustment amount per unit time, and time to reach a steady state for each backup device under different adjustment ranges.

[0015] As a further improvement, the system sends control commands to the corresponding backup equipment according to the backup equipment and power regulation parameters pre-specified in the scheduling strategy. This includes: obtaining the total power regulation amount and completion time limit required by the power grid in response to the power grid dispatch command; determining the combination of backup equipment participating in the regulation and the power regulation task of each equipment based on the total power regulation amount and completion time limit, as well as the power regulation amount per unit time of each backup equipment; determining the start time of each backup equipment based on the start delay time and the regulation time to reach the target power, so that the power regulation periods of each backup equipment are connected to each other on the time axis to meet the completion time limit; and sending control commands to each backup equipment according to the determined start time, power regulation task, and power regulation parameters.

[0016] Secondly, this application also discloses a virtual power plant distributed energy resource optimization scheduling system, comprising: a strategy pre-configuration module for pre-defining multiple combined abnormal states, each combined abnormal state corresponding to a concurrent state in which multiple devices in the distributed energy resource simultaneously malfunction; pre-configuring a corresponding scheduling strategy for each combined abnormal state, wherein the scheduling strategy pre-specifies the standby devices to be called and the power regulation parameters of each standby device; a status monitoring and analysis module for real-time collection of operating status data of the distributed energy resource and determining the current operating status characteristics based on the operating status data; an anomaly matching module for comparing the current operating status characteristics with the pre-defined combined abnormal states to determine the target combined abnormal state corresponding to the currently occurring concurrent state; and a scheduling execution module for executing the scheduling strategy corresponding to the target combined abnormal state and sending control commands to the corresponding standby devices according to the standby devices and power regulation parameters pre-specified in the scheduling strategy. Beneficial effects

[0017] This application discloses a virtual power plant distributed energy resource optimization scheduling method. By pre-defining multiple combined abnormal states and pre-configuring corresponding scheduling strategies for each state, it achieves rapid and accurate response to multiple concurrent abnormal situations of distributed energy resources. The method collects operational status data in real time, determines the current operational status characteristics, and compares them with pre-defined combined abnormal states to identify the target combined abnormal state. Subsequently, the system executes the scheduling strategy corresponding to the target combined abnormal state, sending control commands to the corresponding backup equipment according to pre-specified backup equipment and power regulation parameters. Compared with existing technologies, this application significantly shortens the response time through pre-configured scheduling strategies. When an abnormality occurs, the system does not need to perform time-consuming global optimization calculations and distributed queries, but directly matches the pre-defined abnormal state and executes the corresponding strategy. This effectively solves the problems of response delay and inability to effectively compensate for power gaps within the short time required by the grid in existing technologies, avoiding scheduling task failures and economic penalties. Furthermore, by explicitly specifying the backup equipment and power regulation parameters to be invoked, this method improves the accuracy and operability of scheduling commands, avoiding the problems of fragmented information aggregation and complex calculations. Therefore, this application can significantly improve the dispatch efficiency and reliability of virtual power plants in the face of complex and abnormal situations, ensure the stability of power grid operation, and enhance the reputation of virtual power plants as a reliable regulation resource for the power grid. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the steps of the virtual power plant distributed energy resource optimization scheduling method disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the virtual power plant distributed energy resource optimization scheduling system disclosed in an embodiment of the present invention. Detailed Implementation

[0019] The implementation details of the technical solution in this embodiment are described in detail below: In modern power systems, virtual power plants, as an innovative energy management model, primarily aim to aggregate and manage dispersed, small-scale energy resources, such as photovoltaic power generation equipment, energy storage devices, and production line equipment and auxiliary power equipment with adjustable power consumption, to improve grid stability and energy efficiency. However, in actual operation, virtual power plants face the challenge of efficiently aggregating dispersed energy resources and rapidly responding to grid demands. Traditional energy management systems struggle to flexibly coordinate unstable energy sources like photovoltaic power generation equipment, potentially leading to grid instability and increased energy transaction costs.

[0020] Specifically, when a virtual power plant encounters multiple concurrent unexpected events, such as unforeseen extreme weather causing a sudden failure in photovoltaic power generation, while critical controllable load resources (such as energy storage devices or production line equipment) abruptly and without warning exit the response due to their own commercial reasons, the existing forecasting and planning-based scheduling model will fail. After detecting a significant deviation between power generation and planned values, the system typically activates an automatic compensation mechanism, attempting to cope by recalculating the global optimization.

[0021] However, this approach of recalculating for optimization presents significant technical challenges. It requires communicating with numerous dispersed and heterogeneous secondary resources (such as auxiliary equipment) to obtain their real-time operating status and adjustable potential. Since each subsystem has different communication protocols and response speeds, information aggregation itself is time-consuming. The feedback information may also carry various constraints, requiring the control system to perform a complex recalculation upon receiving this fragmented information in order to piece together a feasible combination from the remaining resources to fill the power gap. The entire process of problem detection, re-querying, recalculation, and issuing new instructions is excessively time-consuming, leading to response delays and an inability to effectively fill the power gap within the grid's required timeframe. Ultimately, this results in scheduling task failure, economic penalties, and damage to its reputation as a reliable regulating resource for the grid.

[0022] To address the aforementioned problems, this application proposes a method for optimal scheduling of distributed energy resources in a virtual power plant, such as... Figure 1 ,include: S101, predefine multiple combined abnormal states, each combined abnormal state corresponding to a concurrent state in which multiple devices in the distributed energy resources simultaneously malfunction; pre-configure a corresponding scheduling strategy for each combined abnormal state, the scheduling strategy pre-specifying the backup devices to be called and the power adjustment parameters of each backup device; the distributed energy resources include photovoltaic power generation equipment, energy storage equipment, production line equipment and auxiliary power equipment; S102, collect the operating status data of distributed energy resources in real time, and determine the current operating status characteristics based on the operating status data; S103, compare the current running state characteristics with the predefined combined abnormal states to determine the target combined abnormal state corresponding to the current concurrent state; S104, execute the scheduling strategy corresponding to the abnormal state of the target combination, and send control commands to the corresponding backup equipment according to the backup equipment and power adjustment parameters specified in the scheduling strategy.

[0023] This application enables rapid and accurate response to complex concurrent anomalies in distributed energy resources of virtual power plants by pre-defining combined abnormal states and configuring corresponding scheduling strategies. It effectively solves the problems of slow response and low efficiency in traditional scheduling modes, thereby improving the operational stability and reliability of virtual power plants.

[0024] To better understand the technical solution proposed in this application, some key terms are explained first. A virtual power plant refers to a new power coordination management model that aggregates dispersed, small-scale distributed energy resources (such as photovoltaic power generation equipment, energy storage equipment, and adjustable loads) as a whole to participate in the power market and grid operation management through advanced information and communication technologies and software systems. Distributed energy resources, in this application, specifically refer to various energy devices managed by the virtual power plant, including photovoltaic power generation equipment (such as solar panel arrays), energy storage equipment (such as battery energy storage systems), production line equipment (such as high-energy-consuming equipment in industrial production, whose power load is adjustable), and auxiliary power equipment (such as central air conditioning systems, cooling pumps, and lighting systems). A combined abnormal state refers to a concurrent state in which multiple devices in a distributed energy resource simultaneously experience abnormalities, such as a sudden drop in photovoltaic power generation and a failure of energy storage equipment occurring simultaneously. A dispatch strategy is a pre-defined set of response plans for a specific combined abnormal state, which clearly specifies the types of backup equipment to be called up, the power adjustment parameters of each backup equipment (such as power reduction and power increase), and the priority order of sending control commands. Operational status data refers to various operational parameters collected in real time from distributed energy resources, such as the output power of photovoltaic power generation equipment, the charging and discharging status of energy storage equipment, and the power load of production line equipment. Operational status characteristics are key information extracted from operational status data that characterizes the current operating status of the equipment, such as characteristics of declining power generation, disconnection of energy storage equipment, and increase in load.

[0025] The core of the virtual power plant distributed energy resource optimization scheduling method proposed in this application lies in achieving rapid response to complex abnormal situations through predefined and matching mechanisms.

[0026] Specifically, in the pre-definition of multiple combined abnormal states, expert experience or data mining methods can be used to identify and define these states. For example, by analyzing historical operating data and fault records, high-frequency, high-impact concurrent abnormal patterns can be identified. Each combined abnormal state corresponds to a concurrent state in which multiple devices in the distributed energy resource simultaneously malfunction. For example, a combined abnormal state can be defined as a sudden drop in the output power of photovoltaic power generation equipment and an interruption of the charging and discharging control connection of energy storage equipment. For each defined combined abnormal state, a corresponding scheduling strategy needs to be pre-configured. The configuration of the scheduling strategy can be formulated by experienced dispatchers based on historical data and simulation results, or optimized and generated through machine learning algorithms. In the scheduling strategy, the backup equipment to be called and the power adjustment parameters of each backup equipment are pre-specified. For example, when photovoltaic power generation drops sharply, the scheduling strategy can specify to call the central air conditioning system in the auxiliary power equipment to reduce its power and set a specific reduction amount. Distributed energy resources can include photovoltaic power generation equipment, energy storage equipment, production line equipment, and auxiliary power equipment. These devices are connected to the virtual power plant management system through corresponding communication interfaces for data acquisition and control command issuance.

[0027] Real-time acquisition of operational status data for distributed energy resources can be achieved by deploying sensors and data acquisition units at each device. For example, the inverter of a photovoltaic power generation device can upload its output power data in real time, the battery management system of an energy storage device can upload its charging / discharging status and connection status, and the PLC (Programmable Logic Controller) of the production line equipment can upload its electricity load data. This data is aggregated into the central management system of the virtual power plant through an Internet of Things (IoT) platform or a SCADA (Supervisory and Data Acquisition) system. Based on this operational status data, the current operational status characteristics can be determined. For example, by calculating the rate of change of photovoltaic output power, it can be determined whether there is a decline in power generation; by monitoring the communication link status of energy storage devices, it can be determined whether there is a disconnection of energy storage devices; and by analyzing the electricity load curve of the production line equipment, it can be determined whether there is a load increase.

[0028] In the step of comparing the current operating state characteristics with predefined combined abnormal states, pattern recognition or similarity matching algorithms can be used. For example, the currently extracted operating state feature vector can be compared with the predefined combined abnormal state feature vector, and the similarity between them can be calculated. Through this comparison, the target combined abnormal state corresponding to the currently occurring concurrent state can be determined. For example, if the current detection shows both a decrease in power generation and a disconnection of energy storage equipment, it can be matched to the predefined combined abnormal state of a sudden drop in photovoltaic power generation output and an interruption of the charging and discharging control connection of energy storage equipment.

[0029] In the execution phase of the scheduling strategy corresponding to the abnormal state of the target combination, once the abnormal state of the target combination is determined, the system will immediately invoke the pre-configured scheduling strategy for that state. According to the pre-specified standby equipment and power adjustment parameters in the scheduling strategy, control commands are sent to the corresponding standby equipment. For example, if the scheduling strategy specifies that the central air conditioning system should reduce its power by 50kW, the system will send the corresponding command to the central air conditioning controller, instructing it to complete the power reduction within a specified time.

[0030] The virtual power plant distributed energy resource optimization scheduling method proposed in this application aims to solve the problems of slow response and low efficiency in traditional scheduling modes when facing multiple concurrent anomalies. Its working principle is to pre-define the possible complex anomalies in a pattern and pre-set an optimized response strategy for each pattern, thereby transforming the real-time complex decision-making process into rapid pattern matching and strategy execution.

[0031] Specifically, during the operation of the distributed energy resources in the virtual power plant, its status monitoring and analysis module continuously collects real-time operational status data from various equipment, including photovoltaic power generation equipment, energy storage equipment, production line equipment, and auxiliary power equipment. This data is used to extract current operational status characteristics, such as abnormal drops in photovoltaic power generation, communication interruptions in energy storage equipment, and sudden increases in production line load. The extraction of these characteristics is real-time, quickly reflecting the latest operational status of the equipment.

[0032] Meanwhile, the strategy pre-configuration module has predefined several combined abnormal states. Each combined abnormal state represents a specific concurrent scenario in which multiple devices in a distributed energy resource simultaneously malfunction. For example, a combined abnormal state might be described as a sudden drop in photovoltaic power generation and an interruption of the charging and discharging control connection of energy storage devices. For each predefined combined abnormal state, a corresponding scheduling strategy is pre-configured. These scheduling strategies are carefully designed and optimized, explicitly specifying the types of backup devices to be invoked under specific abnormal scenarios (such as central air conditioning systems and cooling pumps in auxiliary power equipment) and the specific power adjustment parameters of each backup device (such as power reduction and power increase).

[0033] Once the status monitoring and analysis module determines the current operating status characteristics, the anomaly matching module quickly compares these characteristics with predefined combined anomaly states. This comparison process is highly efficient, aiming to quickly identify which predefined combined anomaly state the current concurrent state most closely resembles, thereby determining the target combined anomaly state. Once the target combined anomaly state is determined, the scheduling execution module immediately starts. It invokes the scheduling strategy pre-bound to the target combined anomaly state and sends control commands to the corresponding backup equipment according to the backup equipment and its power regulation parameters pre-specified in the strategy. For example, if the target combined anomaly state is a sudden drop in photovoltaic power generation and an increase in production line load, and the corresponding scheduling strategy specifies that the cooling pump in the auxiliary power equipment should be de-energized, then the scheduling execution module will send corresponding control commands to the cooling pump to complete the power regulation within a specified time.

[0034] The virtual power plant distributed energy resource optimization scheduling method proposed in this application has significant innovation and advantages over existing technologies in dealing with multiple concurrent anomalies in virtual power plants. Traditional virtual power plant scheduling models typically rely on real-time global optimization calculations to cope with complex concurrent anomalies such as sudden failures in photovoltaic power generation equipment and unannounced shutdowns of energy storage devices or production line equipment. This approach requires communication queries to a large number of dispersed and heterogeneous secondary resources to obtain their real-time operating status and adjustable potential. Due to differences in communication protocols and response speeds, information aggregation is time-consuming, and subsequent complex calculations further extend the response time, making it impossible to effectively compensate for power gaps within the short time required by the grid.

[0035] The core innovation of this application lies in the introduction of the concepts of combined abnormal states and pre-configured scheduling strategies. By pre-defining multiple combined abnormal states, each state corresponds to a concurrent state in which multiple devices in a distributed energy resource simultaneously malfunction. A corresponding scheduling strategy is pre-configured for each combined abnormal state, explicitly specifying the backup devices to be invoked and their power regulation parameters. This method moves the complex decision-making process forward, transforming the traditional problem discovery-real-time query-real-time calculation-instruction issuance model into a problem discovery-pattern matching-strategy execution model.

[0036] For example, in the traditional model, when photovoltaic power generation drops sharply and energy storage devices fail, the system needs to query the real-time status of all adjustable loads and backup power sources one by one, and then perform complex optimization calculations to determine the optimal dispatch scheme. The entire process may take several minutes or even longer, resulting in grid fluctuations that cannot be smoothed out in a timely manner. However, using the method of this application, once the combined abnormal state of a sharp drop in photovoltaic power generation and energy storage device failure is detected, the system will immediately invoke the pre-configured dispatch strategy, such as directly instructing the central air conditioning system and cooling pumps in auxiliary power equipment to reduce their power. The entire process can be completed within seconds, greatly improving response efficiency and grid stability.

[0037] In some embodiments of this application, the predefined combined abnormal states of the above-mentioned virtual power plant distributed energy resource optimization scheduling method may include the following types: First combined abnormal state: the output power decline rate of the photovoltaic power generation equipment exceeds a first preset threshold, and the charging and discharging control connection of the energy storage equipment is interrupted; Second combined abnormal state: the output power decline rate of the photovoltaic power generation equipment exceeds a first preset threshold, and the increase in the power load of the production line equipment exceeds a second preset threshold.

[0038] The first abnormal state describes a scenario where both the power generation and energy storage sides experience anomalies simultaneously. Specifically, when the power generation capacity of photovoltaic power generation equipment drops sharply due to weather changes, equipment failures, or other reasons, and its output power decline rate exceeds a preset first threshold, the charging and discharging control connection of the energy storage equipment is simultaneously interrupted, causing the energy storage system to be unable to charge and discharge energy normally. The concurrent occurrence of these two abnormal situations will significantly impact the stable operation of the virtual power plant and the power supply reliability of the power grid.

[0039] Furthermore, the second combined abnormal state describes a scenario where anomalies on the generation side and the load side occur simultaneously. Specifically, when the rate of decline in the output power of photovoltaic power generation equipment exceeds a first preset threshold, it indicates insufficient power generation capacity; simultaneously, when the increase in the electricity load of the production line equipment exceeds a second preset threshold, it means a sharp rise in electricity demand. The simultaneous occurrence of these two situations will lead to a severe supply-demand imbalance for the virtual power plant, potentially causing voltage or frequency fluctuations in the local power grid, and even affecting the normal operation of the production line.

[0040] This application enables the system to accurately identify complex concurrent failure modes that may occur in distributed energy resources by predefining these specific combined abnormal states. Traditional methods may only focus on the abnormality of a single device, ignoring the more serious impact of the abnormality of multiple devices working together. By clearly defining the first combined abnormal state and the second combined abnormal state, the system can match the real-time collected operating status characteristics with these predefined complex scenarios, thereby more accurately determining the type and degree of risk currently faced.

[0041] Specifically, determining the current operating status characteristics based on the operating status data may include the following steps: monitoring the output power decline rate of the photovoltaic power generation equipment, and generating a power decline characteristic when the output power decline rate exceeds the first preset threshold; monitoring the charging and discharging control connection status of the energy storage equipment, and generating an energy storage equipment disconnection characteristic when the charging and discharging control connection is interrupted; monitoring the change in the power load of the production line equipment, and generating a load increase characteristic when the increase in power load exceeds the second preset threshold.

[0042] The monitoring of the output power decline rate of photovoltaic power generation equipment aims to obtain the real-time operating status of photovoltaic power generation units. When the output power decline rate exceeds a preset first threshold, it indicates that the photovoltaic power generation may be abnormal, such as being affected by sudden weather changes (e.g., cloud cover, heavy rainfall) or equipment failure. At this time, the system will generate a power decline feature to mark this abnormal situation.

[0043] Furthermore, monitoring the charging and discharging control connection status of energy storage devices is to ensure the normal communication link between the energy storage system and the virtual power plant control center. When the charging and discharging control connection is interrupted, it means that the energy storage device may be unable to receive or execute scheduling commands, thus losing its regulation capability. At this time, the system will generate an energy storage device disconnection feature to indicate the unavailability of the energy storage resource.

[0044] In addition, monitoring changes in the electrical load of production line equipment aims to understand the real-time demand for industrial load. When the increase in electrical load exceeds a preset second threshold, it indicates that the production line may be experiencing a sudden surge in load demand or abnormal load fluctuations. At this time, the system will generate a load increase characteristic to reflect the abnormal situation on the load side.

[0045] The solution proposed in this application enables real-time and accurate detection of various abnormal operating states by performing refined monitoring of key equipment in distributed energy resources. By setting specific thresholds (such as a first preset threshold and a second preset threshold) and state judgment conditions (such as interruption of charging / discharging control connection), the system can transform complex raw operating data into operating state characteristics with clear physical meaning (such as power generation decline, energy storage device disconnection, and load increase). The generation of these characteristics provides standardized and structured input for subsequent comparison of the current operating state characteristics with predefined combinations of abnormal states, thereby ensuring the accuracy and efficiency of abnormal state identification.

[0046] Specifically, in some embodiments of the above-mentioned virtual power plant distributed energy resource optimization scheduling method, the scheduling strategy pre-specifies the standby equipment to be called and the power adjustment parameters of each standby equipment, including: determining the type of standby equipment to be called, wherein the standby equipment is the auxiliary power equipment, including at least one of central air conditioning system, cooling pump, lighting system or non-core production line; determining the power reduction amount or power increase amount of each standby equipment; and determining the priority order of sending control commands to each standby equipment.

[0047] Specifically, in the distributed energy resource optimization and scheduling method of virtual power plants, when a scheduling strategy needs to be executed, it is first necessary to determine which equipment can be used as backup equipment. These backup devices are limited to auxiliary power equipment, characterized by their fact that they typically do not directly participate in the core production process; therefore, adjusting their power has a relatively small impact on overall production, offering high flexibility and controllability. Auxiliary power equipment can include, but is not limited to, at least one of the following: central air conditioning systems, cooling pumps, lighting systems, or non-core production lines. For example, central air conditioning systems and cooling pumps can change their electrical load by adjusting their operating power; lighting systems can reduce electricity consumption by dimming or partially shutting down; and non-core production lines can adjust the operating status of some equipment to achieve power regulation without affecting the output of the main products.

[0048] Furthermore, for the identified standby equipment, its specific power regulation parameters need to be defined. These parameters include the power reduction or increase amount for each standby device. Power reduction refers to the amount of electricity load that the standby equipment can reduce in response to dispatch instructions. For example, when the grid needs peak shaving and valley filling or when distributed energy generation is insufficient, the overall load can be reduced by decreasing the power of auxiliary equipment. Power increase refers to the amount of electricity load that the standby equipment can increase in response to dispatch instructions. For example, when distributed energy generation is excessive, the excess electricity can be absorbed by increasing the power of auxiliary equipment. These regulation amounts are typically determined based on the equipment's rated power, current operating status, and preset regulation range.

[0049] Furthermore, to ensure the effective execution of scheduling instructions and the stability of system operation, it is necessary to determine the priority order for sending control instructions to each backup device. This priority order can be determined based on various factors, such as the device's response speed, adjustment accuracy, degree of impact on production, and current device availability. For example, auxiliary power equipment with the least impact on production and the fastest response speed can be prioritized to achieve the required power regulation effect in the shortest time, while maximizing the continuity of core production.

[0050] This application's solution, by clearly defining the specific types of backup equipment and their power regulation parameters in the dispatch strategy, enables the virtual power plant to selectively choose appropriate auxiliary power equipment to respond to abnormal conditions of distributed energy resources. By pre-designating auxiliary power equipment as backup resources, unnecessary interference with core production processes can be effectively avoided, thereby achieving flexible dispatch of distributed energy resources while ensuring production continuity. Simultaneously, by determining the amount of power reduction or increase, and the priority of control commands, the execution of dispatch commands becomes more refined and efficient, enabling precise power adjustment based on actual needs and optimizing resource allocation during the adjustment process.

[0051] In some embodiments of this application, the current running state characteristics are compared with predefined combined abnormal states to determine the target combined abnormal state corresponding to the currently occurring concurrent state. However, in its implementation, if only simple feature comparison is used, it may not accurately identify the combined abnormal state that best matches the current situation, especially when multiple abnormal state characteristics overlap or certain characteristics have a more critical impact on scheduling decisions, which may lead to misjudgment or low scheduling efficiency. To address this, this application further proposes a scheme to optimize the above comparison process by quantifying feature similarity and introducing an importance coefficient to more accurately determine the target combined abnormal state.

[0052] The steps described above, which compare the current running state features with predefined combined abnormal states to determine the target combined abnormal state corresponding to the current concurrent state, specifically include: calculating the similarity value between the current running state features and the corresponding features in the combined abnormal state; configuring an importance coefficient for each feature in the combined abnormal state; weighting and summing the similarity values ​​according to the importance coefficients to obtain a matching value; sorting the matching values ​​of each combined abnormal state; and determining the combined abnormal state with the highest matching value that exceeds a preset threshold as the target combined abnormal state.

[0053] Specifically, the current operating status characteristics refer to the power generation decline characteristics, energy storage device disconnection characteristics, or load increase characteristics generated by monitoring data such as the output power decline rate of photovoltaic power generation equipment, the charging and discharging control connection status of energy storage equipment, and the changes in the power load of production line equipment. Combined abnormal states predefine concurrent states where multiple devices in distributed energy resources simultaneously experience abnormalities, such as a first combined abnormal state or a second combined abnormal state. The similarity value calculated between the current operating status characteristics and the corresponding characteristics in the combined abnormal states aims to quantify the degree of agreement between the current actual operating state and each predefined combined abnormal state at the feature level. This similarity value can be calculated in various ways, such as binary matching (1 if the feature exists, 0 if it does not), fuzzy logic matching, or distance metrics based on numerical differences (such as Euclidean distance or cosine similarity). Its purpose is to provide a quantitative basis for subsequent matching degree calculations.

[0054] Furthermore, importance coefficients are assigned to each feature in the combined abnormal states to reflect the relative importance of different features in judging the abnormal state. For example, some features may be directly related to power grid stability or production safety, and their importance coefficients should be set to higher values; while other features may have a smaller impact, and their importance coefficients can be set to lower values. These importance coefficients can be based on expert experience, historical data analysis, or trained and optimized through machine learning algorithms. The similarity values ​​are weighted and summed according to the importance coefficients to obtain the matching degree value. This matching degree value comprehensively considers the similarity and importance of each feature, and can more comprehensively and objectively reflect the degree of matching between the current operating state and each combined abnormal state. In practical applications, sorting the matching degree values ​​of each combined abnormal state can clearly identify the combined abnormal state closest to the current operating state. Subsequently, the combined abnormal state with the highest matching degree value exceeding a preset threshold is determined as the target combined abnormal state. The setting of this preset threshold aims to ensure that the determined target combined abnormal state not only has the highest matching degree, but also that its matching degree reaches a sufficiently high confidence level, avoiding the forced selection of an optimal but inaccurate abnormal state when the matching degree is generally low.

[0055] This application's solution effectively addresses the inaccuracy issues that may arise from traditional simple comparisons by introducing similarity values, importance coefficients, and matching scores. Specifically, the similarity value calculation quantifies the degree of agreement between the current operating state and predefined combinations of abnormal states, avoiding the limitations of binary judgment. The importance coefficient configuration allows the system to differentiate the critical impact of different features on virtual power plant operation and scheduling decisions, ensuring that key abnormal features receive higher weight during the matching process. By weighted summing of similarity values, the resulting matching score more comprehensively and precisely reflects the degree of agreement between the current concurrent state and various predefined combinations of abnormal states. Finally, by sorting the matching scores and filtering them using preset thresholds, the system ensures that the identified target combinations of abnormal states not only have the highest matching score but also achieve the reliability required for practical applications, thereby avoiding scheduling errors or inefficiencies caused by misjudgments.

[0056] The following is a specific example to illustrate this. Assume that the real-time operational status data collected by the virtual power plant, after processing, determines that the current operational status characteristics include a decrease in power generation and a disconnection of energy storage devices, while a load increase characteristic is not generated. Simultaneously, the system predefines two combined abnormal states: First combined abnormal state: The output power decrease rate of the photovoltaic power generation equipment exceeds a first preset threshold, and the charging and discharging control connection of the energy storage device is interrupted. Second combined abnormal state: The output power decrease rate of the photovoltaic power generation equipment exceeds the first preset threshold, and the increase in the electrical load of the production line equipment exceeds a second preset threshold.

[0057] To determine the target combined abnormal state, the similarity value between the current running state features and the corresponding features in the two combined abnormal states is first calculated.

[0058] For the first combination of abnormal states: The similarity value between the power generation decline characteristic and the rate of decline of the output power of photovoltaic power generation equipment exceeding the first preset threshold is 1.0 (perfect match).

[0059] The similarity between the disconnection feature of the energy storage device and the interruption of the charge and discharge control connection of the energy storage device is 1.0 (exact match).

[0060] For the second combination of abnormal states: The similarity value between the power generation decline characteristic and the rate of decline of the output power of photovoltaic power generation equipment exceeding the first preset threshold is 1.0 (perfect match).

[0061] The similarity between the load increase feature and the increase in electrical load of production line equipment exceeding the second preset threshold is 0.0 (not a match, because no load increase feature has been generated at present).

[0062] Next, importance coefficients are assigned to each feature. Assume the importance coefficient for the power generation decline feature is 0.4, the importance coefficient for the energy storage device disconnection feature is 0.3, and the importance coefficient for the load increase feature is 0.3. Then, the similarity values ​​are weighted and summed according to the importance coefficients to obtain the matching degree value: The matching degree value of the first combination of abnormal states = (1.0 * 0.4) + (1.0 * 0.3) = 0.7.

[0063] The matching degree value of the second combination of abnormal states = (1.0 * 0.4) + (0.0 * 0.3) = 0.4.

[0064] Finally, the matching scores of each combined abnormal state are sorted. In this example, the matching score of the first combined abnormal state (0.7) is higher than that of the second combined abnormal state (0.4). Assuming the preset threshold is 0.6, since the matching score of the first combined abnormal state (0.7) is higher than the preset threshold of 0.6 and is the highest value, the first combined abnormal state is determined as the target combined abnormal state corresponding to the currently occurring concurrent state. The system will then execute the scheduling strategy corresponding to the first combined abnormal state.

[0065] This application further proposes that before executing the scheduling strategy corresponding to the above-mentioned target combination abnormal state, it also includes: querying the enterprise management system for equipment availability information through the application programming interface; calculating the actual available adjustable power of each standby device based on the equipment availability information, and correcting the standby devices and their power adjustment parameters specified in the scheduling strategy accordingly.

[0066] Specifically, an application programming interface (API) can be understood as a type of software middleware that allows data exchange and function calls between different software systems. Here, this API is used by the virtual power plant scheduling system to communicate with the enterprise's internal production management system or equipment management system to obtain real-time equipment status data. Equipment availability information refers to data obtained from the enterprise management system regarding the current status of standby equipment, such as whether the equipment is running, under maintenance, its current load, adjustable power range, and whether there are any production tasks restricting its participation in scheduling. Calculating the actual adjustable power of each standby device refers to assessing the maximum power reduction or increase that each standby device can provide at the current moment based on the retrieved equipment availability information. For example, if a central air conditioning system is providing cooling for a critical production area, its adjustable power may be limited to avoid impacting production. Correcting the standby devices and their power adjustment parameters specified in the scheduling strategy refers to dynamically adjusting the pre-configured scheduling strategy based on the calculated actual adjustable power. This may include removing currently unavailable standby devices from the scheduling list or adjusting the power adjustment parameters of a standby device to match its actual adjustable capacity.

[0067] This application's solution effectively solves the problem of mismatch between pre-configured strategies and real-time operating status by introducing equipment availability query and strategy correction steps before executing scheduling strategies. Specifically, by obtaining equipment availability information from the enterprise management system in real time through an application programming interface, the virtual power plant scheduling system can accurately understand the current operating status and adjustability potential of each standby device. Based on this real-time information, the system can calculate the actual available adjustable power of each standby device, thereby avoiding sending invalid instructions to unavailable or limited-capacity devices. Thus, the pre-configured scheduling strategy can be dynamically corrected according to the actual situation, ensuring the effectiveness and feasibility of scheduling instructions and avoiding scheduling failures or negative impacts on enterprise production caused by information lag.

[0068] In some preferred embodiments, a specific example is given below. Assume a virtual power plant manages distributed energy resources in an industrial park, including photovoltaic power generation equipment, energy storage equipment, and multiple production line equipment and auxiliary electrical equipment, such as a central air conditioning system, cooling pumps, and lighting systems. When the system detects that the output power decline rate of the photovoltaic power generation equipment exceeds a first preset threshold, and the increase in the power load of the production line equipment exceeds a second preset threshold, the system identifies a second combined abnormal state and prepares to execute a pre-configured scheduling strategy. This pre-configured strategy may specify that the central air conditioning system and cooling pumps should be reduced in power to balance the load. However, before executing this strategy, the system first queries the park's enterprise management system via an application programming interface for the real-time availability information of these auxiliary electrical equipment. The enterprise management system returns information showing that the central air conditioning system is providing precise temperature control for a critical production workshop, and its power reduction is strictly limited, while a cooling pump is undergoing planned maintenance and is currently unavailable. Based on this real-time availability information, the system calculates that the actual available adjustable power of the central air conditioning system is far lower than the preset value, and the actual available adjustable power of the cooling pump is zero. Therefore, the system will revise its original scheduling strategy, removing the cooling pump from the call list and adjusting the power reduction of the central air conditioning system to its actual manageable range. Simultaneously, the system may reassess, based on the revised strategy, whether other backup equipment (such as the lighting system) needs to be called to compensate for the power shortfall. Finally, following the revised scheduling strategy, the system sends control commands to the available backup equipment that meets the required capacity, thus ensuring the effective execution of the scheduling task while avoiding disruption to critical production processes.

[0069] In some embodiments described above, equipment availability information is queried from the enterprise management system via an application programming interface (API), and the backup equipment and its power regulation parameters specified in the scheduling strategy are modified accordingly. However, in actual implementation, relying solely on the equipment availability information provided by the enterprise management system may not be sufficient to fully assess the actual adjustability of backup equipment under the current production state and its potential impact on core production processes. For example, although some auxiliary electrical equipment may appear available at the system level, its power regulation under specific production loads or process conditions may cause unacceptable interference or safety risks to critical production links. If these problems are not addressed, the modified scheduling strategy may still have execution risks, failing to ensure scheduling security and production stability.

[0070] In response, the revised scheduling strategy further includes: sending a simulated adjustment instruction to the production impact verification module deployed at each backup device, the simulated adjustment instruction including the simulated adjustment range; receiving the safety verification result returned by the production impact verification module, the safety verification result including the safe adjustable power range and safety impact level of the backup device in the current production state, the safety impact level being determined by querying the coupling relationship between the backup device and the core production process; and removing backup devices with a safe adjustable power range of zero or a safety impact level exceeding a preset level from the scope of the scheduling strategy.

[0071] Specifically, the Production Impact Verification (PIV) module refers to software or hardware units deployed at or closely associated with each standby device. Its main function is to assess in real-time the potential impact of power adjustments to standby devices on current production activities. This module can acquire operational data from standby devices and their associated production processes and analyze it based on preset logic or models. A simulated adjustment command is a virtual, non-actually executed adjustment command sent by the system to the PIV module before actual control commands are sent. It includes the planned power adjustment amount for the standby device, such as a power reduction or increase. The simulated adjustment range refers to the specific power adjustment amount included in the simulated adjustment command. Receiving the safety verification result means that after receiving the simulated adjustment command, the PIV module generates and returns an evaluation report based on its internal logic and real-time production data. The safe adjustable power range refers to the upper and lower limits of the power adjustment that the standby device can undergo in its current state without affecting production safety and the normal operation of core processes. The safety impact level is a quantitative assessment of the severity of the impact that power regulation of standby equipment may have on production. It can be categorized into several levels, such as no impact, minor impact, moderate impact, and severe impact. This safety impact level is determined by querying the coupling relationship between the standby equipment and the core production process, i.e., the degree of correlation and dependency between them. This coupling relationship can be pre-configured in the system, for example, defined through equipment topology diagrams, process flow diagrams, or expert knowledge bases. Removing standby equipment with a safe adjustable power range of zero or a safety impact level exceeding a preset level from the scope of the scheduling strategy means that if a standby device is completely unadjustable, or if its adjustment may have an unacceptable impact on production, that device will be removed from the current scheduling strategy and will no longer be considered for power regulation.

[0072] This application's solution introduces a production impact verification module, adding a crucial pre-verification step after the scheduling strategy is revised but before the actual execution of scheduling instructions. Specifically, the system no longer relies solely on general availability information provided by the enterprise management system; instead, it sends simulated adjustment instructions to the production impact verification modules deployed at each backup device. This module, combining the current operating status of the backup device with its coupling relationship to the core production process, performs real-time safety and production impact assessments of the simulated adjustment instructions, returning the safe adjustable power range and safety impact level of the backup device under the current production state. Thus, the scheduling system can perform secondary screening of backup devices based on more refined and real-time production impact assessment results. Backup devices with a safe adjustable power range of zero or a safety impact level exceeding a preset level will be removed from the scope of the scheduling strategy, even if they appear available in the enterprise management system. This mechanism ensures that the final executed scheduling strategy not only considers equipment availability but also fully considers production safety and the stability of core business operations, avoiding unnecessary interference or risks to production caused by blindly scheduling auxiliary equipment.

[0073] In some preferred embodiments, suppose a virtual power plant needs to respond to a power reduction command from the power grid, and according to a query in the enterprise management system, a central air conditioning system on a production line is identified as a callable backup device, theoretically capable of reducing its power by 50kW. However, before executing the scheduling strategy, the system sends a simulated 50kW reduction command to the production impact verification module deployed at the central air conditioning system. This verification module, considering the high-precision product processing task currently underway on the production line (which has strict requirements for ambient temperature and humidity), assesses that reducing the central air conditioning system's power by 50kW will lead to an increase in workshop temperature, thus affecting product quality. Its safety impact level is determined to be moderate, exceeding the preset slight impact tolerance level. Simultaneously, the module may calculate that the maximum safe adjustable power range of the central air conditioning system without affecting production is only 10kW. Based on this safety verification result, the scheduling system will revise the original scheduling strategy, adjusting the actual adjustable power of the central air conditioning system to 10kW, or, if its safety impact level is too high, directly removing it from the scope of this scheduling strategy and seeking other backup devices for adjustment. In this way, even under emergency dispatch needs, the core interests of industrial production can be ensured to remain unharmed.

[0074] This application further proposes that the above-mentioned scheduling strategy also includes power response timing parameters for each backup device. The power response timing parameters include the start-up delay time, power adjustment amount per unit time, and duration to reach a steady state for the backup device under different adjustment ranges.

[0075] Specifically, power response timing parameters refer to a set of parameters used to describe the dynamic characteristics of standby equipment's power output or consumption over time after receiving a dispatch command. These parameters are crucial for achieving refined and time-coordinated dispatching. Among them, the start-up delay time refers to the time interval between the standby equipment receiving the control command and actually starting power regulation. This delay may be caused by factors such as the equipment's own physical inertia, control system response time, or communication delay. The power regulation per unit time refers to the amount of power that the standby equipment can change per unit time, i.e., the rate at which its power ramps up or decreases. This parameter reflects the equipment's regulation speed and flexibility. The time to reach a steady state refers to the time required for the standby equipment to reach its target power output or consumption and remain stable from the start of regulation. This parameter comprehensively considers the equipment's start-up delay and regulation rate, and is a key indicator for evaluating the overall response capability of the equipment. In practical applications, these power response timing parameters can be pre-calibrated or estimated in real time based on the type, model, operating conditions, and historical data of the standby equipment, and stored in the dispatching strategy for retrieval when formulating a dispatching plan.

[0076] The proposed solution introduces power response timing parameters into the scheduling strategy, enabling the scheduling system to more comprehensively and accurately grasp the dynamic response capabilities of each backup device. When the scheduling strategy needs to be executed, the system not only knows which backup devices to call and the total power they need to adjust, but also accurately calculates the optimal start-up time, adjustment rate, and duration for each backup device based on these timing parameters. For example, for devices with longer start-up delays, start-up commands can be sent in advance; for devices with slower adjustment rates, longer adjustment time can be allocated or they can be coordinated with other devices for adjustment. Therefore, the issuance of scheduling commands will be more closely aligned with the actual physical characteristics of the backup devices, thereby ensuring that distributed energy resources can adjust power according to the expected timing and magnitude.

[0077] In some preferred embodiments, it is assumed that the virtual power plant needs to cope with a combined abnormal state of a sharp drop in the output power of photovoltaic power generation equipment and a sudden increase in the power load of production line equipment. In this case, the dispatching system needs to quickly call upon auxiliary power equipment to reduce power consumption to balance supply and demand. When formulating the dispatching strategy, the system queries the power response timing parameters of each auxiliary power equipment. For example, the central air conditioning system may have a long start-up delay time (e.g., 10 seconds) and a moderate power adjustment per unit time (e.g., 50 kW / min), while the cooling pump may have a short start-up delay time (e.g., 3 seconds) and a fast power adjustment per unit time (e.g., 100 kW / min). When the system needs to achieve a total power reduction of 200 kW within a short time limit (e.g., 2 minutes), the dispatching strategy will precisely plan the start-up time and adjustment curves of the central air conditioning system and the cooling pump based on these parameters. For example, the system may prioritize sending control commands to the cooling pump, causing it to begin rapidly reducing power after 3 seconds; simultaneously, it may send commands to the central air conditioning system in advance, causing it to begin reducing power at its inherent rate after a 10-second delay. This coordination ensures that the two devices can work together to reduce the power by 200kW within a 2-minute time limit, thus effectively responding to abnormal situations and maintaining system stability.

[0078] This application further proposes the following steps for sending control commands to corresponding backup equipment according to the backup equipment and power regulation parameters pre-specified in the scheduling strategy: responding to the grid dispatch command to obtain the total power regulation amount and completion time limit required by the grid; determining the combination of backup equipment participating in the regulation and the power regulation task of each equipment based on the total power regulation amount and completion time limit, and the power regulation amount per unit time of each backup equipment; determining the start time of each backup equipment based on the start delay time of each backup equipment and the regulation time to reach the target power, so that the power regulation periods of each backup equipment are connected to each other on the time axis to meet the completion time limit; and sending control commands to each backup equipment according to the determined start time, power regulation task, and power regulation parameters.

[0079] Specifically, responding to grid dispatch instructions by obtaining the total power regulation required by the grid and the completion deadline refers to the virtual power plant system receiving and parsing dispatch instructions issued by the grid in real time through communication with the grid dispatch center, such as via an application programming interface (API) or a dedicated communication protocol. This instruction typically specifies the total power regulation required from the virtual power plant (e.g., how many megawatts of power to increase or decrease) and the deadline for completing this regulation task. The total power regulation can be understood as the grid's adjustment requirement for the overall power output or consumption of the virtual power plant within a specific time period, aiming to maintain grid frequency or voltage stability; the completion deadline specifies the time point when the regulation task must be completed, aiming to ensure the timeliness and effectiveness of grid dispatch.

[0080] The process of determining the combination of backup equipment participating in the regulation and the power regulation task of each equipment, based on the total power regulation amount and completion time limit, and the power regulation amount per unit time of each backup equipment, refers to the system performing optimization calculations after receiving the grid dispatch instruction. This calculation is based on the currently available backup equipment list, the power regulation amount per unit time of each backup equipment (i.e., the amount of power it can regulate per unit time), and its maximum adjustable power. The aim of this calculation is to select an optimal combination of backup equipment and assign a specific power regulation task to each backup equipment in the combination, ensuring that the sum of the regulation amounts of all equipment can meet the total power regulation requirement of the grid. This process may involve multi-objective optimization algorithms, such as minimizing regulation costs or maximizing regulation efficiency while meeting the total regulation requirement.

[0081] In practical applications, the startup time of each backup device is determined based on its startup delay time and the adjustment time required to reach the target power. This ensures that the power adjustment periods of each backup device are seamlessly connected on the timeline to meet the completion deadline. This means that after determining the backup device combination and their respective adjustment tasks, the system further utilizes preset power response timing parameters in the scheduling strategy. Specifically, for each selected backup device, the system considers the startup delay time required from receiving the instruction to starting actual power adjustment, and the adjustment time required from the start of adjustment to reaching its target power. By accurately calculating these time parameters, the system can plan an optimal startup time for each backup device, allowing the power adjustment processes of different devices to be smoothly connected on the timeline, avoiding adjustment interruptions or overlaps. This ensures that the overall power adjustment curve of the virtual power plant can continuously and stably reach the total power adjustment required by the grid and be completed within the specified completion deadline.

[0082] This application's solution obtains the specific total power regulation amount and completion time limit by responding to grid dispatch instructions, thus providing a clear target for the dispatch of the virtual power plant. Based on this, by comprehensively considering the power regulation amount per unit time of each standby device, it can intelligently select appropriate standby device combinations and allocate regulation tasks to ensure the achievement of the total regulation amount. Furthermore, by accurately calculating the start-up delay time of each standby device and the regulation time to reach the target power, it can precisely determine the start-up time of each device, enabling seamless connection of the power regulation processes of different devices on the time axis. This ensures that the virtual power plant can function as a whole, smoothly and efficiently completing the grid's power regulation requirements within the specified time limit, effectively solving the problems of untimely response and uncoordinated regulation that may exist in traditional dispatching.

[0083] In some preferred embodiments, a specific example is given below. Assume a grid dispatch command requires a virtual power plant to provide a 5MW power reduction within 8 minutes. The virtual power plant's currently available backup equipment includes auxiliary equipment A and auxiliary equipment B. Auxiliary equipment A has a start-up delay of 1 minute, a power reduction rate of 1MW / minute, and a maximum adjustable power of 3MW; auxiliary equipment B has a start-up delay of 0.5 minutes, a power reduction rate of 2MW / minute, and a maximum adjustable power of 4MW.

[0084] First, the system responds to the grid dispatch command, obtaining a total power regulation capacity of 5MW and a completion time limit of 8 minutes. Second, based on the unit-time power regulation capacity and maximum adjustable power of each standby device, the system determines to call auxiliary equipment A and auxiliary equipment B. To meet the 5MW total requirement, the system can allocate a power regulation task of 2MW to auxiliary equipment A and a power regulation task of 3MW to auxiliary equipment B. Next, the system determines the start-up time based on the power response timing parameters of each device. For auxiliary equipment A, its start-up delay time is 1 minute, and regulating 2MW requires 2 minutes (2MW / 1MW / minute). Therefore, auxiliary equipment A requires 1 + 2 = 3 minutes from start-up to completion of regulation. For auxiliary equipment B, its start-up delay time is 0.5 minutes, and regulating 3MW requires 1.5 minutes (3MW / 2MW / minute). Therefore, auxiliary equipment B requires 0.5 + 1.5 = 2 minutes from start-up to completion of regulation. To ensure seamless power regulation periods and meet the 8-minute completion time limit, the system calculates the following startup times: Auxiliary equipment B starts at t=0. It completes startup at t=0.5 minutes and regulates 3MW of power between t=0.5 minutes and t=2 minutes. Auxiliary equipment A starts at t=0.5 minutes. It completes startup at t=1.5 minutes and regulates 2MW of power between t=1.5 minutes and t=3.5 minutes. Thus, during t=0.5 minutes to t=1.5 minutes, auxiliary equipment B performs power regulation; from t=1.5 minutes onwards, auxiliary equipment A and auxiliary equipment B simultaneously perform power regulation until auxiliary equipment B completes its regulation task at t=2 minutes, and auxiliary equipment A continues regulation until t=3.5 minutes. The entire 5MW power regulation is completed at t=3.5 minutes, far below the 8-minute completion time limit, and the regulation process is smoothly connected. Finally, the system sends corresponding control commands to auxiliary electrical equipment A and auxiliary electrical equipment B according to the determined start time, power regulation task, and power regulation parameters.

[0085] Secondly, this application proposes a virtual power plant distributed energy resource optimization scheduling system, such as... Figure 2,include: The strategy pre-configuration module 201 is used to predefine multiple combined abnormal states, each combined abnormal state corresponding to a concurrent state in which multiple devices in the distributed energy resources simultaneously malfunction; and to pre-configure a corresponding scheduling strategy for each combined abnormal state, wherein the scheduling strategy pre-specifies the backup devices to be called and the power adjustment parameters of each backup device. The status monitoring and analysis module 202 is used to collect the operating status data of distributed energy resources in real time and determine the current operating status characteristics based on the operating status data; The anomaly matching module 203 is used to compare the current running state characteristics with predefined combined anomaly states to determine the target combined anomaly state corresponding to the currently occurring concurrent state. The scheduling execution module 204 is used to execute the scheduling strategy corresponding to the abnormal state of the target combination, and send control commands to the corresponding backup equipment according to the backup equipment and power adjustment parameters specified in the scheduling strategy.

[0086] The virtual power plant distributed energy resource optimization and scheduling system proposed in this application, through its modular design, pre-defines and solidifies complex anomaly handling logic, enabling rapid and accurate response to multiple concurrent anomalies in the face of distributed energy resources in the virtual power plant. The strategy pre-configuration module is responsible for pre-defining and storing potential concurrent anomaly scenarios and their corresponding solutions; the status monitoring and analysis module perceives the system's operating status in real time and extracts key features; the anomaly matching module quickly identifies the current anomaly pattern; and finally, the scheduling execution module rapidly issues control commands based on the preset strategy. Thus, this system effectively solves the problems of slow response and low efficiency in traditional scheduling modes, significantly improving the operational stability and reliability of the virtual power plant.

[0087] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for optimal scheduling of distributed energy resources in a virtual power plant, characterized in that, include: Multiple combined abnormal states are predefined, each corresponding to a concurrent state in which multiple devices in the distributed energy resources simultaneously malfunction; a corresponding scheduling strategy is pre-configured for each combined abnormal state, the scheduling strategy pre-specifying the backup devices to be called and the power adjustment parameters of each backup device; the distributed energy resources include photovoltaic power generation equipment, energy storage equipment, production line equipment and auxiliary power equipment; Real-time acquisition of operational status data of distributed energy resources, and determination of current operational status characteristics based on the operational status data; The current operating state characteristics are compared with predefined combined abnormal states to determine the target combined abnormal state corresponding to the current concurrent state. The scheduling strategy corresponding to the abnormal state of the target combination is executed, and control commands are sent to the corresponding backup devices according to the backup devices and power adjustment parameters specified in the scheduling strategy.

2. The virtual power plant distributed energy resource optimization scheduling method according to claim 1, characterized in that, The combined abnormal states include: First abnormal state combination: The output power of the photovoltaic power generation equipment decreases at a rate exceeding a first preset threshold, and the charging and discharging control connection of the energy storage equipment is interrupted; Second combination of abnormal states: The rate of decrease in the output power of the photovoltaic power generation equipment exceeds the first preset threshold, and the increase in the power load of the production line equipment exceeds the second preset threshold.

3. The virtual power plant distributed energy resource optimization scheduling method according to claim 1, characterized in that, Determining the current operating status characteristics based on the operating status data includes: The rate of decrease in output power of the photovoltaic power generation equipment is monitored, and when the rate of decrease in output power exceeds a first preset threshold, a power generation decrease characteristic is generated; Monitor the charging and discharging control connection status of the energy storage device, and generate a disconnection feature when the charging and discharging control connection is interrupted; The power load of the production line equipment is monitored, and when the increase in power load exceeds a second preset threshold, a load increase feature is generated.

4. The virtual power plant distributed energy resource optimization scheduling method according to claim 1, characterized in that, The scheduling strategy pre-specifies the backup equipment to be called up and the power regulation parameters of each backup equipment, including: The type of backup equipment to be called up is determined, wherein the backup equipment is the auxiliary electrical equipment, including at least one of the following: central air conditioning system, cooling pump, lighting system or non-core production line; Determine the amount of power reduction or increase for each standby device; Determine the priority order for sending control commands to each backup device.

5. The virtual power plant distributed energy resource optimization scheduling method according to claim 3, characterized in that, The step of comparing the current operating state characteristics with predefined combined abnormal states to determine the target combined abnormal state corresponding to the currently occurring concurrent state includes: Calculate the similarity value between the current operating state features and the corresponding features in the combined abnormal states; An importance coefficient is assigned to each feature in the combined abnormal state, and the similarity value is weighted and summed according to the importance coefficient to obtain the matching degree value; The matching degree values ​​of each combination of abnormal states are sorted, and the combination of abnormal states with the highest matching degree value that exceeds the preset threshold is determined as the target combination of abnormal states.

6. The method for optimized scheduling of distributed energy resources in a virtual power plant according to claim 1, characterized in that, Before executing the scheduling strategy corresponding to the target combined abnormal state, the following steps are also included: Query device availability information from the enterprise management system via application programming interface; Based on the equipment availability information, the actual available adjustable power of each standby device is calculated, and the standby devices and their power adjustment parameters specified in the scheduling strategy are corrected accordingly.

7. The virtual power plant distributed energy resource optimization scheduling method according to claim 6, characterized in that, After modifying the backup equipment and its power regulation parameters specified in the scheduling strategy, the method further includes: Send a simulation adjustment command to the production impact verification module deployed at each backup device, the simulation adjustment command including the simulation adjustment range; Receive the safety verification result returned by the production impact verification module. The safety verification result includes the safe adjustable power range and safety impact level of the backup equipment in the current production state. The safety impact level is determined by querying the coupling relationship between the backup equipment and the core production process. Backup devices with a safe adjustable power range of zero or a safety impact level exceeding a preset level will be removed from the scope of the scheduling strategy.

8. The method for optimized scheduling of distributed energy resources in a virtual power plant according to claim 1, characterized in that, The scheduling strategy also includes power response timing parameters for each backup device, which include the startup delay time, power adjustment amount per unit time, and time to reach a steady state for the backup device under different adjustment ranges.

9. The method for optimal scheduling of distributed energy resources in a virtual power plant according to claim 8, characterized in that, The step of sending control commands to the corresponding backup equipment according to the backup equipment and power regulation parameters pre-specified in the scheduling strategy includes: In response to grid dispatch instructions, obtain the total power regulation required by the grid and the completion time limit; Based on the total power regulation amount and completion time limit, as well as the power regulation amount per unit time of each standby device, determine the combination of standby devices participating in the regulation and the power regulation task of each device; Based on the start-up delay time of each backup device and the adjustment time to reach the target power, the start-up time of each backup device is determined so that the power adjustment periods of each backup device are connected on the time axis to meet the completion time limit. Control commands are sent to each standby device according to the determined start time, power regulation task, and power regulation parameters.

10. A virtual power plant distributed energy resource optimization scheduling system, characterized in that, include: The strategy pre-configuration module is used to predefine multiple combined abnormal states, each of which corresponds to a concurrent state in which multiple devices in the distributed energy resources simultaneously malfunction; and to pre-configure a corresponding scheduling strategy for each combined abnormal state, wherein the scheduling strategy pre-specifies the backup devices to be called and the power adjustment parameters of each backup device. The status monitoring and analysis module is used to collect real-time operational status data of distributed energy resources and determine the current operational status characteristics based on the operational status data. The anomaly matching module is used to compare the current running state characteristics with predefined combined anomaly states to determine the target combined anomaly state corresponding to the currently occurring concurrent state. The scheduling execution module is used to execute the scheduling strategy corresponding to the abnormal state of the target combination, and send control commands to the corresponding backup devices according to the backup devices and power adjustment parameters specified in the scheduling strategy.

Images