Green electricity direct connection oriented multi-garden heterogeneous resource collaborative scheduling method and system

Abstract

The application relates to a multi-garden heterogeneous resource collaborative scheduling method and system for green electricity direct connection, comprising: uniformly modeling different types of heterogeneous resources of multiple gardens to obtain corresponding regulation resource models of each resource; based on the regulation resource model, a multi-objective collaborative optimization model and corresponding safety constraint conditions are constructed; under the safety constraint conditions, the multi-objective collaborative optimization model is decomposed into a mutual aid and grid exchange sub-problem between gardens and an internal optimization sub-problem of each garden by using an alternating direction multiplier method and is iteratively solved until a preset convergence condition is met, so that a target collaborative scheduling strategy is obtained; and the heterogeneous resources of the multiple gardens are collaboratively scheduled based on the target collaborative scheduling strategy. According to the scheme, the power operation cost of each garden can be reduced and the operation stability of the power system can be improved by uniformly modeling the multi-garden heterogeneous power resources, constructing the multi-objective collaborative optimization model and performing distributed solving.

Description

Technical Field

[0001] This application relates to the field of power system technology, and in particular to a method and system for collaborative scheduling of heterogeneous resources in multiple parks for direct green power connection. Background Technology

[0002] Electricity resources in industrial parks can include various types of renewable energy, energy storage devices, and demand response resources. However, the instability and intermittency of distributed energy devices in different industrial parks can pose certain challenges to the stable operation of the power system, especially in high-energy-consuming scenarios such as coal mining areas, coal chemical industrial parks, and computing centers.

[0003] In related technologies, green electricity direct connection technology can be used to directly connect new energy power generation devices to end users in industrial parks, thereby reducing transmission losses and improving the efficiency of green electricity utilization. This reduces the impact of different distributed energy devices in industrial parks on the power system at different times. However, there will be significant temporal and spatial mismatch between green electricity supply and demand in different parks. For example, photovoltaic systems built in coal mining subsidence areas have surplus electricity during the midday peak output, while coal chemical parks require a large amount of green electricity due to continuous production at night. Due to the lack of a cross-park surplus and shortage mutual assistance mechanism, some parks abandon electricity while others purchase electricity at high prices. In addition, the different energy-consuming devices in different parks lead to different load characteristics in different parks, making it difficult and costly to allocate power resources between different parks, and making it impossible to ensure the safe and stable operation of the power system. Summary of the Invention

[0004] To address or partially address the problems existing in related technologies, this application provides a method and system for collaborative scheduling of heterogeneous resources in multiple parks for direct green power connections. By uniformly modeling heterogeneous power resources in multiple parks, constructing a multi-objective collaborative optimization model, and using an alternating direction multiplier algorithm to achieve distributed solution, it can reduce the power operation costs of each park and improve the frequency security and operational stability of the power system.

[0005] The first aspect of this application provides a multi-park heterogeneous resource collaborative scheduling method for direct green power connection, applied to a source-grid-load-storage collaborative control system. The method includes: collecting real-time operation data and forecast data from multiple parks; uniformly modeling different types of heterogeneous resources to obtain regulation resource models corresponding to each resource; the heterogeneous resources include at least impact loads in coal mining areas, continuous adjustable loads in coal chemical industrial parks, interruptible loads in computing centers, and energy storage systems; all regulation resource models are uniformly associated with the regulation amount and regulation cost of the heterogeneous resources; based on the regulation resource models, a multi-objective collaborative optimization model and corresponding security constraints are constructed; wherein the multi-objective collaborative optimization model aims to minimize the weighted sum of total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits. The safety constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints that are transformed into linear inequalities through analytical processing. Under these safety constraints, the multi-objective collaborative optimization model is decomposed into inter-park mutual assistance and grid exchange sub-problems and intra-park optimization sub-problems using the alternating direction multiplier method. These sub-problems are solved iteratively by exchanging boundary information until convergence, yielding candidate scheduling strategies. The candidate scheduling strategies undergo online frequency safety pre-evaluation and correction. If the frequency index exceeds the safety threshold, a virtual inertia penalty term is returned to the intra-park optimization sub-problem, and the solution is iterated again until all frequency indexes meet the requirements, yielding the target collaborative scheduling strategy. Based on the target collaborative scheduling strategy, heterogeneous resources across multiple parks are collaboratively scheduled.

[0006] In conjunction with the first aspect, in one possible implementation of the first aspect, the unified modeling of different types of heterogeneous resources to obtain the corresponding regulating resource models for each resource includes: modeling the impact load of the coal mining area based on translation time window constraints and total charging energy conservation constraints to obtain a corresponding first regulating resource model; wherein, the regulating amount of the first regulating resource model is the regulating amount of the impact load of the coal mining area relative to the reference power; and modeling the continuous adjustable load of the coal chemical industrial park based on power operating range constraints and regulation rate constraints to obtain a corresponding second regulating resource model; wherein, the The adjustment amount of the second adjustment resource is the adjustment amount of the continuous adjustable load of the coal chemical industrial park relative to the reference power; based on the interruption power constraint and the interruption duration constraint, the interruptible load of the computing center is modeled to obtain the corresponding third adjustment resource model; wherein, the adjustment amount of the third adjustment resource model is the adjustment amount of the interruption power corresponding to the interruptible load of the computing center; based on the charge and discharge power range constraint and the dynamic state of charge constraint, the energy storage system is modeled to obtain the corresponding fourth adjustment resource model; wherein, the adjustment amount of the fourth adjustment resource model is the adjustment amount of the charge and discharge power of the energy storage system.

[0007] In conjunction with the first aspect, in one possible implementation of the first aspect, the construction of frequency stability constraints based on the adjustment resource model, which are transformed into linear inequalities through analytical processing, includes: calculating the virtual inertia time constant of each park in each time period according to the ratio of grid-type equipment to grid-type equipment in each park, the operating status of the energy storage system, and the power of the tie line; summing the virtual inertia time constants of each park to obtain the total system inertia of the source-grid-load-storage coordinated control system in each time period; and constructing linear inequality constraints related to the total system inertia based on the total system inertia, the adjustment resource model, the system frequency change rate, and the minimum frequency deviation.

[0008] In conjunction with the first aspect, in one possible implementation of the first aspect, the step of constructing a linear inequality constraint related to the total system inertia based on the total system inertia, the adjustment resource model, the maximum power change value, and the minimum frequency deviation value includes: obtaining the power disturbance amount when the source-grid-load-storage coordinated control system experiences a preset maximum power disturbance; obtaining a first expression corresponding to the system frequency change rate and a second expression corresponding to the minimum frequency deviation of the source-grid-load-storage coordinated control system based on the power disturbance amount, the system damping ratio, the natural oscillation frequency, and the total system inertia; wherein the system frequency change rate is less than a first safety threshold, and the minimum frequency deviation is less than a second safety threshold; and constructing a linear inequality constraint related to the total system inertia based on the first expression and the second expression.

[0009] In conjunction with the first aspect, in one possible implementation of the first aspect, the step of performing online pre-evaluation and correction of the candidate scheduling strategy for frequency security, and if the frequency index exceeds the security threshold, returning a virtual inertia penalty term to the internal optimization sub-problem of the park, and iterating again until all frequency indexes meet the requirements to obtain the target cooperative scheduling strategy, includes: inputting the current total system inertia and current total adjustment power corresponding to the current candidate scheduling strategy into the average system frequency reduction model to obtain the current system frequency change rate, current minimum point frequency deviation, and current quasi-steady-state frequency deviation corresponding to the current candidate scheduling strategy; if the current system frequency change rate exceeds the first security threshold, and / or the current minimum point frequency deviation exceeds the second security threshold, and / or the current quasi-steady-state frequency deviation exceeds the third security threshold, then generating the virtual inertia penalty term; returning the virtual inertia penalty term to the internal optimization sub-problem of the park, and iterating again until all frequency indexes meet the requirements to obtain the target cooperative scheduling strategy.

[0010] In conjunction with the first aspect, in one possible implementation of the first aspect, the construction of a multi-objective collaborative optimization model based on the adjusted resource model includes: constructing a total system operating cost function based on time-of-use electricity purchase price, time-of-use electricity sales price, the exchange power between each park and the target grid, and the adjustment cost of adjustable resources within each park; constructing a curtailment penalty cost function based on the curtailment penalty unit price and the curtailment volume of each park; constructing a green revenue function based on the actual green electricity consumption and green electricity subsidy price; and performing a weighted summation of the total system operating cost function, the curtailment penalty cost function, and the green revenue function to construct a multi-objective collaborative optimization model with the goal of minimizing the weighted sum; wherein the multi-objective collaborative optimization model is positively correlated with the total system operating cost function and the curtailment penalty cost function, and negatively correlated with the green revenue function.

[0011] In conjunction with the first aspect, in one possible implementation of the first aspect, under the security constraints, the multi-objective collaborative optimization model is decomposed into an inter-park mutual assistance and grid exchange sub-problem and an internal optimization sub-problem within each park using the alternating direction multiplier method. The solution is obtained by iteratively exchanging boundary information until convergence, yielding candidate scheduling strategies. This includes: in the inter-park mutual assistance and grid exchange sub-problem, determining the adjustment amount of heterogeneous resources corresponding to each park as a fixed value and solving for it to obtain the optimal mutual assistance power between each park and the optimal exchange power between each park and the grid when the sum of the electricity purchase cost and mutual assistance cost is minimized; in the internal optimization sub-problem of each park, the adjustment amount of heterogeneous resources corresponding to each park is determined as a fixed value and solved to obtain the optimal mutual assistance power between each park and the optimal exchange power between each park and the grid; in the internal optimization sub-problem of each park, the adjustment amount of heterogeneous resources corresponding to each park is determined as a fixed value and solved to obtain the optimal mutual assistance power between each park and the optimal exchange power between each park and the grid when the sum of the electricity purchase cost and mutual assistance cost is minimized; in the internal optimization sub-problem of each park, the adjustment amount of heterogeneous resources corresponding to each park is determined as a fixed value and solved to obtain the optimal mutual assistance ... In the sub-problem of optimization within the industrial park, the optimal mutual assistance power and the optimal exchange power are determined as fixed values. The local optimization objective is to minimize the adjustment cost within each industrial park. The optimal adjustment amount of the heterogeneous resources corresponding to each industrial park is solved separately, and the solution is obtained by iteratively exchanging boundary information until convergence, thus obtaining the candidate scheduling strategy. In the iterative solution process, Lagrange multipliers are used to exchange boundary information between the sub-problem of mutual assistance and grid exchange between industrial parks and the sub-problem of optimization within each industrial park. The boundary information includes the marginal price of mutual assistance power and the net load curve after adjustment within each industrial park, until the convergence condition is met.

[0012] The second aspect of this application provides a multi-park heterogeneous resource collaborative scheduling system for direct green power connection, comprising: an acquisition module for collecting real-time operation data and forecast data from multiple parks, uniformly modeling different types of heterogeneous resources, and obtaining adjustment resource models corresponding to each resource; the heterogeneous resources include at least impact loads in coal mining areas, continuous adjustable loads in coal chemical industrial parks, interruptible loads in computing centers, and energy storage systems, and all the adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources; and a construction module for constructing a multi-objective collaborative optimization model and corresponding security constraints based on the adjustment resource models; wherein the multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits, and the security constraints include at least power consumption... The system includes: a rate balance constraint, energy storage operation constraint, tie-line capacity constraint, resource-specific operation constraint, and frequency stability constraint transformed into a linear inequality through analytical processing; a processing module, used to decompose the multi-objective collaborative optimization model into inter-park mutual assistance and grid exchange sub-problems and intra-park optimization sub-problems under the aforementioned safety constraints using the alternating direction multiplier method, and solves the problem until convergence by iteratively exchanging boundary information to obtain candidate scheduling strategies; a correction module, used to perform online pre-evaluation and correction of the candidate scheduling strategies for frequency safety, and if the frequency index exceeds the safety threshold, returns a virtual inertia penalty term to the intra-park optimization sub-problem, and iterates again until all frequency indexes meet the requirements to obtain the target collaborative scheduling strategy; and a collaborative scheduling module, used to perform collaborative scheduling of heterogeneous resources in multiple parks based on the target collaborative scheduling strategy.

[0013] A third aspect of this application provides an electronic device, comprising: Processor; and A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.

[0014] A fourth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method described above.

[0015] The technical solution provided in this application may include the following beneficial effects: This application discloses a method and system for collaborative scheduling of heterogeneous resources across multiple industrial parks for direct green power connections. This method is applied to a source-grid-load-storage collaborative control system, comprising: collecting real-time operating data and forecast data from multiple industrial parks; uniformly modeling different types of heterogeneous resources to obtain corresponding adjustment resource models for each resource; the heterogeneous resources include at least impact loads from coal mining areas, continuous adjustable loads from coal chemical industrial parks, interruptible loads from computing centers, and energy storage systems; all adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources; based on the adjustment resource models, a multi-objective collaborative optimization model and corresponding safety constraints are constructed; wherein, the multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits, while ensuring safety... The constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints transformed into linear inequalities through analytical processing. Under safety constraints, the alternating direction multiplier method is used to decompose the multi-objective collaborative optimization model into inter-park mutual assistance and grid exchange sub-problems, and intra-park optimization sub-problems. These sub-problems are solved iteratively by exchanging boundary information until convergence, yielding candidate scheduling strategies. Online frequency safety pre-evaluation and correction are performed on the candidate scheduling strategies. If the frequency index exceeds the safety threshold, a virtual inertia penalty term is returned to the intra-park optimization sub-problem, and the solution is iterated again until all frequency indexes meet the requirements, yielding the target collaborative scheduling strategy. Based on the target collaborative scheduling strategy, heterogeneous resources across multiple parks are collaboratively scheduled. This approach, by uniformly modeling heterogeneous power resources across multiple parks, constructing a multi-objective collaborative optimization model, and employing the alternating direction multiplier algorithm for distributed solution, can reduce the power operation costs of each park and improve the frequency safety and operational stability of the power system.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0018] Figure 1 This is a flowchart illustrating a multi-park heterogeneous resource collaborative scheduling method for direct green electricity connection, as shown in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a multi-park heterogeneous resource collaborative scheduling system for direct green electricity connection, as shown in an embodiment of this application. Figure 3 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application. Detailed Implementation

[0019] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0020] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0022] Electricity resources in industrial parks can include various types of renewable energy, energy storage devices, and demand response resources. However, the instability and intermittency of distributed energy devices in different industrial parks can pose certain challenges to the stable operation of the power system, especially in high-energy-consuming scenarios such as coal mining areas, coal chemical industrial parks, and computing centers.

[0023] In related technologies, green electricity direct connection technology can be used to directly connect new energy power generation devices to end users in industrial parks, thereby reducing transmission losses and improving the efficiency of green electricity utilization. This reduces the impact of different distributed energy devices in industrial parks on the power system at different times. However, there will be significant temporal and spatial mismatch between green electricity supply and demand in different parks. For example, photovoltaic systems built in coal mining subsidence areas have surplus electricity during the midday peak output, while coal chemical parks require a large amount of green electricity due to continuous production at night. Due to the lack of a cross-park surplus and shortage mutual assistance mechanism, some parks abandon electricity while others purchase electricity at high prices. In addition, the different energy-consuming devices in different parks lead to different load characteristics in different parks, making it difficult and costly to allocate power resources between different parks, and making it impossible to ensure the safe and stable operation of the power system.

[0024] To address the aforementioned issues, this application provides a method for collaborative scheduling of heterogeneous resources across multiple parks for direct green power connections. This method can uniformly model heterogeneous power resources across multiple parks, construct a multi-objective collaborative optimization model, and use an alternating direction multiplier algorithm to achieve distributed solution. This can reduce the power operation costs of each park and improve the frequency security and operational stability of the power system.

[0025] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0026] Figure 1 This is a flowchart illustrating a multi-park heterogeneous resource collaborative scheduling method for direct green electricity connection, as shown in an embodiment of this application.

[0027] See Figure 1 A collaborative scheduling method for heterogeneous resources in multiple parks for direct green power connections is applied to a source-grid-load-storage collaborative control system, including: S110: Collect real-time operation data and forecast data from multiple parks, and perform unified modeling of different types of heterogeneous resources to obtain the corresponding adjustment resource models for each resource; heterogeneous resources include at least the impact load of coal mining areas, the continuous adjustable load of coal chemical industrial parks, the interruptible load of computing centers, and energy storage systems. All adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of heterogeneous resources.

[0028] Specifically, the source-grid-load-storage coordinated control system is a comprehensive control platform that integrates power sources (such as photovoltaic and wind power), power grids (including public power grids and microgrids within the park), power loads (various electrical devices), and energy storage devices (such as battery energy storage). It can allocate different types of power resources in multiple parks. Heterogeneous resources are power resources with different operating characteristics and adjustment space in different parks. The source-grid-load-storage coordinated control system can acquire real-time operating data and forecast data from different parks in real time, and obtain different types of heterogeneous resources from them. Then, it models the heterogeneous resources. This adjustment resource model can be used to reflect the adjustment amount and adjustment cost of heterogeneous resources. The adjustment amount and adjustment cost of the adjustment resource model corresponding to different heterogeneous resources can be different.

[0029] Specifically, impact loads in coal mining areas refer to load resources with large changes in power demand within a short period of time. They are characterized by instantaneous high power consumption and generally have a fixed total power consumption. However, the loads can be shifted as a whole during a certain period of time within an allowable time window. For example, the charging of electric mining trucks in coal mining areas requires the electric mining trucks to be fully charged within a certain period of time (total power consumption is fixed), while the specific start and end times of charging can be adjusted according to the actual situation.

[0030] Specifically, continuous adjustable loads in coal chemical industrial parks are loads whose power demand can be continuously and smoothly adjusted within a certain range. Continuous adjustable loads usually have certain adjustment rate limitations and adjustment costs, such as compressors, carbon capture devices, or electrolyzers in coal chemical industrial parks. The energy consumption of electrolyzers and carbon capture devices is linearly related to hydrogen production and CO2 capture.

[0031] Specifically, interruptible loads in computing centers refer to power loads that can be temporarily interrupted under specific conditions, and their corresponding adjustment amounts and adjustment costs can be set according to actual needs.

[0032] Specifically, energy storage resources are devices that store electrical energy and release it when needed, such as battery energy storage systems in industrial parks. Energy storage resources have parameters such as charge and discharge power and state of charge (SOC).

[0033] S120: Based on the resource adjustment model, a multi-objective collaborative optimization model and corresponding security constraints are constructed. The multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost and green benefits. The security constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints that are converted into linear inequalities through analytical processing.

[0034] Specifically, the total operating cost of the system is the total operating cost of the park within a certain scheduling cycle, including factors such as electricity purchase, electricity sales, energy storage charging and discharging losses, and internal resource adjustment. The curtailment penalty cost is the economic cost incurred when green electricity (such as photovoltaic and wind power) generated in the park is not fully absorbed and curtailed. The green electricity absorption benefit is the economic benefit obtained by the park through effective absorption of green electricity. The carbon emission cost is the cost required for the park to emit carbon. By constructing a multi-objective collaborative optimization model, the corresponding power resource scheduling scheme can be obtained by adjusting the adjustment amount of each park to minimize the total operating cost of the entire system.

[0035] Specifically, each power zone must meet safety constraints during operation and regulation. These constraints include power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints. Power balance constraints require that, at any given time, the total power supply within the zone (e.g., generation, energy storage discharge, power purchase, and mutual assistance input) must equal the total power demand. Power demand can be generated from load consumption, energy storage charging, power sales, and mutual assistance output. Energy storage operation constraints define thresholds that each power resource within the zone, such as generators, energy storage, and adjustable loads, cannot exceed during operation, such as maximum / minimum power, regulation rate, and state of charge range. Tie-line capacity constraints refer to the maximum power transmission limit that transmission lines connecting different zones or zones to the external power grid can bear. Frequency stability constraints require that the operating frequency of the power system remain within a preset safe range.

[0036] S130: Under safety constraints, the multi-objective collaborative optimization model is decomposed into inter-park mutual assistance and power grid exchange sub-problems and internal optimization sub-problems of each park by using the alternating direction multiplier method. The solution is obtained by iteratively exchanging boundary information until convergence, thus obtaining candidate scheduling strategies.

[0037] Specifically, the alternating direction multiplier algorithm decomposes a multi-objective collaborative optimization model into sub-problems of inter-park mutual assistance and grid exchange, and internal optimization sub-problems within each park. It achieves convergence of the global optimal solution by exchanging information between sub-problems through multiplier terms. The inter-park mutual assistance and grid exchange sub-problem, in the alternating direction multiplier algorithm, is responsible for coordinating power exchange between parks and power exchange with the external grid. The internal optimization sub-problem, in the alternating direction multiplier algorithm, is a sub-problem where each park independently solves its internal power resource scheduling scheme. By iteratively solving the inter-park mutual assistance and grid exchange sub-problems and the internal optimization sub-problems, and exchanging boundary information, a preset convergence condition is met. For example, when the change in the total operating cost of the power system obtained from two consecutive iterations is less than a preset value, a candidate collaborative scheduling strategy is obtained. This candidate collaborative scheduling strategy includes power exchange schemes between parks and specific resource adjustment schemes within each park.

[0038] S140: Perform online pre-evaluation and correction of frequency security for candidate scheduling strategies. If the frequency index exceeds the security threshold, return the virtual inertia penalty term to the internal optimization sub-problem of the park, and iterate again until all frequency indexes meet the requirements to obtain the target cooperative scheduling strategy.

[0039] Specifically, after obtaining candidate scheduling strategies, frequency security online pre-evaluation and correction can be performed on the candidate scheduling strategies. For example, a preset engineered frequency security pre-evaluation module can be used to pre-evaluate and correct the candidate scheduling strategies. This engineered frequency security pre-evaluation module can perform evaluations through offline simulation. For example, the candidate scheduling strategy can be input into this module and subjected to long-term dynamic simulation to verify frequency stability. If the frequency index is found to exceed the safety threshold, a virtual inertia penalty term is generated and returned to the internal optimization sub-problem of the park. This allows the internal optimization sub-problem to be iteratively solved again until all frequency indexes meet the requirements, thus obtaining the target cooperative scheduling strategy.

[0040] S150: Based on the target-based collaborative scheduling strategy, it performs collaborative scheduling of heterogeneous resources in multiple parks.

[0041] Specifically, based on the power mutual assistance scheme between different parks in the target coordinated scheduling strategy, the power of each park can be coordinated and scheduled. Furthermore, the power resources can be coordinated and scheduled by adjusting the heterogeneous resources of each park. In this way, heterogeneous resources with different response times, adjustment capacities, and cost functions can be uniformly modeled as adjustment resource models associated with adjustment quantities and adjustment costs. Then, a multi-objective coordinated optimization model can be jointly constructed. Based on safety constraints and the alternating direction multiplier algorithm, the solution rate of the multi-objective coordinated optimization model can be improved. Under the condition of minimizing the total operating cost of the power system, the heterogeneous resources between different parks can be adjusted, reducing the waste of heterogeneous resources and improving the operating efficiency and stability of the entire power system.

[0042] For example, the source-grid-load-storage coordinated control system simultaneously controls two industrial parks: Park A (a coal mining area mainly generating photovoltaic power, with excess photovoltaic output at midday and impact loads such as electric mining truck charging demand at night) and Park B (a coal chemical industrial park with continuous demand for green electricity at night and equipment such as electrolytic cells). These two parks are interconnected via direct connection lines and connected to the external power grid. Park A collects real-time photovoltaic power generation and electric mining truck charging status; Park B collects real-time electrolytic cell operating power and production plan. Then, based on preset mapping rules, the electric mining truck charging load in Park A is modeled as an impact load resource with specific adjustment amount and adjustment cost, the electrolytic cells in Park B are modeled as a continuous adjustable load resource with adjustment range and adjustment cost, and the energy storage systems in each park are modeled as energy storage resources with charging and discharging power and loss costs. Next, a collaborative optimization model is constructed based on these resource adjustment models. This model incorporates the electricity purchase and sale costs of Park A, the production adjustment costs of Park B, and the curtailment penalties and green electricity consumption benefits of each park into a unified objective function. At the same time, a series of safety constraints are constructed, such as ensuring that the operation of power resources within each park does not exceed its physical limits; limiting the transmission power of the interconnection line between Park A and Park B to not exceed its capacity; and ensuring that the frequency of the entire multi-park system remains stable within a safe range.Subsequently, under this safety constraint, the collaborative optimization model is decomposed into inter-park mutual assistance optimization sub-problems and intra-park optimization sub-problems based on the alternating direction multiplier algorithm. The inter-park mutual assistance optimization sub-problem coordinates the mutual assistance power between parks A, B, and C, as well as their exchange power with the external power grid, to minimize the overall power purchase cost and mutual assistance cost. The intra-park optimization sub-problems, upon receiving the mutual assistance power and Lagrange multipliers (reflecting the marginal price of mutual assistance power), independently optimize their internal energy storage systems, impact loads, continuously adjustable loads, and interruptible loads of the computing center to minimize local operating costs and meet the corresponding safety constraints of the park. These two sub-problems are solved iteratively by exchanging boundary information until a preset convergence condition is met. For example, when the mutual assistance power and internal adjustment amount between parks change little during continuous iterations. If the preset value is reached, it indicates that the power system has reached a stable state, the iteration stops, and a candidate coordinated dispatch strategy is obtained. Then, the frequency security online pre-evaluation and correction of the candidate coordinated dispatch strategy is performed. For example, the system will calculate the total system inertia at the current moment based on the energy storage operation status and the proportion of grid-type equipment in each park under the candidate dispatch strategy, and evaluate the system's frequency change rate and minimum point frequency deviation in combination with the preset maximum power disturbance. If the evaluation results show that any frequency index exceeds the safety threshold, such as the frequency change rate being too fast or the minimum point frequency deviation being too small, the system will immediately return a virtual inertia penalty term to the optimization sub-problem within each park. This penalty term will increase the cost of insufficient virtual inertia, so that each park will adjust its energy storage charging and discharging scheme when iterating again until all frequency indices meet the safety requirements and the target coordinated dispatch strategy is obtained. Finally, based on this goal-oriented collaborative scheduling strategy, the power resources of each park are coordinated and scheduled. For example, at noon, the photovoltaic output of park A is excessive, and the scheduling strategy instructs park A to transmit green electricity to park B. At the same time, the electric mining trucks in park A delay charging. At night, the electrolyzers in park B increase power consumption and receive green electricity from park A. Through this collaborative scheduling, the overall economic benefits of multiple parks are improved and the utilization rate of green energy is increased.

[0043] In one possible implementation, a unified model is performed for different types of heterogeneous resources to obtain corresponding regulating resource models for each resource. This includes: modeling the impact load of a coal mining area based on translation time window constraints and total charging energy conservation constraints to obtain a corresponding first regulating resource model; wherein the regulating amount of the first regulating resource model is the regulating amount of the impact load relative to the baseline power; modeling the continuous adjustable load of a coal chemical industrial park based on power operating range constraints and regulation rate constraints to obtain a corresponding second regulating resource model; wherein the regulating amount of the second regulating resource model is the regulating amount of the continuous adjustable load of the coal chemical industrial park relative to the baseline power; modeling the interruptible load of a computing center based on interruption power constraints and interruption duration constraints to obtain a corresponding third regulating resource model; wherein the regulating amount of the third regulating resource model is the regulating amount of the interruption power corresponding to the interruptible load of the computing center; and modeling the energy storage system based on charge / discharge power range constraints and dynamic state of charge constraints to obtain a corresponding fourth regulating resource model; wherein the regulating amount of the fourth regulating resource model is the regulating amount of the charge / discharge power of the energy storage system.

[0044] Specifically, the time window shift constraint adjusts the operating time of the impact load in the coal mine area before and after the shift without affecting functionality. The total charging energy conservation constraint ensures that the total energy consumption of the impact load remains unchanged throughout the entire scheduling cycle after the time shift. For example, the first adjustment resource model may include the adjustment amount of the impact load. and adjustment costs , , The first adjustment cost coefficient, The corresponding constraints are , , For load Rated power, For time window, For load The total energy required can be directly obtained from the industrial park by the source-grid-load-storage coordinated control system. , , The corresponding numerical value.

[0045] Specifically, the power operating range constraint is the maximum operating power of the continuously adjustable load in the coal chemical industrial park at any given time, such as the maximum operating power of the electrolyzers in the coal chemical industrial park. The adjustment rate is the rate of change of the continuously adjustable load power, such as the maximum power change rate per minute of the electrolyzers in the coal chemical industrial park. For example, the second adjustment resource model may include the adjustment amount of the continuously adjustable load. and adjustment costs ,in, , and These are the first coefficient and the second coefficient, respectively. The corresponding constraints are as well as ,in, and These are continuous adjustable loads. Minimum permissible operating power and rated power, For continuous adjustable load Basic operating power, For continuous adjustable load The maximum power change per unit time can be directly obtained from the industrial park by the source-grid-load-storage coordinated control system. , , , The corresponding numerical value.

[0046] Specifically, interruptible loads in a computing center can be temporarily interrupted or have their power reduced under certain conditions. The interruption power constraint refers to the maximum amount of power that can be reduced when the interruptible load is interrupted; for example, this can be related to the number of interruptible servers within the computing center, the adjustability of the cooling system load, etc. The interruption duration constraint refers to the maximum duration for which the interruptible load can be interrupted during a single interruption event. The adjustment amount of the interruptible load in the computing center during time period t can be... The interrupt power constraint corresponding to the interruptible load of the computing center can be: , To interrupt the load The maximum interrupt capacity and interrupt duration constraint can be: interrupt duration The adjustment cost corresponding to interruptible load is: , To compensate for the interruption costs.

[0047] Specifically, the charge / discharge power range constraint is used to constrain the maximum charging and discharging power of the energy storage system, while the dynamic state of charge constraint is used to constrain the stored energy (state of charge, SOC) within the energy storage device to remain within the safe operating range (minimum and maximum SOC). For example, the fourth regulation resource model may include the energy storage system. Corresponding adjustment amount and adjustment costs ,in, , This is the second adjustment cost coefficient. The corresponding constraints are , , For energy storage systems State of charge, and energy storage system Charging efficiency and discharging efficiency and energy storage system The maximum charging power and maximum discharging power of the power source, grid, load, and energy storage coordinated control system can be directly obtained from the park. , , , , The corresponding numerical value.

[0048] In one possible implementation, based on the adjustment resource model, frequency stability constraints are constructed by analytically transforming them into linear inequalities. This includes: calculating the virtual inertia time constant of each park in each time period based on the ratio of grid-type equipment to grid-connected equipment in each park, the operating status of the energy storage system, and the power of the tie line; summing the virtual inertia time constants of each park to obtain the total system inertia of the source-grid-load-storage coordinated control system in each time period; and constructing linear inequality constraints related to the total system inertia based on the total system inertia, the adjustment resource model, the system frequency change rate, and the minimum frequency deviation.

[0049] Specifically, grid-connected equipment can be inverters with grid-connecting capabilities, synchronous generators, etc., while grid-connected equipment can be conventional grid-connected inverters, wind turbines, etc. By weighted averaging the ratio of grid-connected to grid-connected equipment, the operating status of the energy storage system, and the power of the tie lines in each park, the virtual inertial time constant of each park at each time period can be obtained. The total inertia of the system can be calculated using the following formula: , The total system inertia is the sum of all virtual inertial time constants of the entire source-grid-load-storage collaborative control system within a specific time period. Finally, a linear inequality constraint related to the total system inertia is constructed using the total system inertia, the adjustment resource model, the system frequency change rate, and the minimum frequency deviation. This frequency stability constraint ensures the frequency stability of the entire multi-park system.

[0050] In one possible implementation, a linear inequality constraint related to the total system inertia is constructed based on the total system inertia, the regulation resource model, the maximum power change value, and the minimum frequency deviation value. This includes: obtaining the power disturbance amount when the source-grid-load-storage coordinated control system experiences a preset maximum power disturbance; obtaining a first expression corresponding to the system frequency change rate and a second expression corresponding to the minimum frequency deviation of the source-grid-load-storage coordinated control system based on the power disturbance amount, the system damping ratio, the natural oscillation frequency, and the total system inertia; wherein the system frequency change rate is less than a first safety threshold, and the minimum frequency deviation is less than a second safety threshold; and constructing a linear inequality constraint related to the total system inertia based on the first and second expressions.

[0051] Specifically, the power disturbance is the power imbalance caused by extreme events such as the tripping of the largest single unit, a sudden increase or decrease in maximum load, etc. The power system can directly obtain this power disturbance, and at the same time obtain the system damping ratio and natural oscillation frequency, to construct the first expression corresponding to the system frequency change rate:

[0052] And the second expression corresponding to the minimum frequency deviation:

[0053] in, The rate of change of frequency; This represents the maximum value of the adjustment amount for heterogeneous resources; In order to be in The total inertia of the system at that time; This is the first safety threshold; Minimum frequency deviation; The system damping ratio; It is the natural oscillation frequency; This is the second safety threshold.

[0054] Specifically, both the first and second safety thresholds can be preset. The frequency change rate is the change in the power grid frequency per unit time. When power imbalance occurs in the grid, the frequency will change. An excessively high frequency change rate may cause generators to disconnect from the grid or protection devices to activate, affecting the stable operation of the grid. Therefore, the frequency change rate needs to be limited to the first safety threshold, for example, 0.5 Hz / s. To determine the maximum power regulation capability that all heterogeneous resources can provide within a specific time period, the total system inertia is defined as the magnitude of the inertia of the entire source-grid-load-storage coordinated control system at time t. A larger inertia results in a slower frequency response to power disturbances and a smaller rate of frequency change. The minimum frequency deviation is the difference between the lowest point the system frequency drops to after a power disturbance and the rated frequency. It reflects the lowest frequency level the system reaches during the frequency drop process. An excessively low minimum frequency may cause system collapse; therefore, the minimum frequency deviation needs to be limited within a second safety threshold, such as 1 Hz or 0.8 Hz, the system damping ratio ζ, and the natural oscillation frequency. All of these can be directly obtained, enabling the power source-grid-load-storage coordinated control system to maintain the frequency within a safe range when dealing with various power disturbances, thereby ensuring the reliability and operational quality of the entire power system.

[0055] In one possible implementation, online pre-evaluation and correction of frequency safety are performed on candidate scheduling strategies. If the frequency indicators exceed the safety threshold, a virtual inertia penalty term is returned to the internal optimization sub-problem of the park, and the solution is iterated again until all frequency indicators meet the requirements to obtain the target cooperative scheduling strategy. This includes: inputting the current total system inertia and current total regulation power corresponding to the current candidate scheduling strategy into the average system frequency reduction model to obtain the current system frequency change rate, current minimum point frequency deviation, and current quasi-steady-state frequency deviation corresponding to the current candidate scheduling strategy; if the current system frequency change rate exceeds the first safety threshold, and / or the current minimum point frequency exceeds the second safety threshold, and / or the current quasi-steady-state frequency deviation exceeds the third safety threshold, a virtual inertia penalty term is generated; the virtual inertia penalty term is returned to the internal optimization sub-problem of the park, and the solution is iterated again until all frequency indicators meet the requirements to obtain the target cooperative scheduling strategy.

[0056] Specifically, during the generation of the collaborative scheduling strategy, the frequency stability of the system can be monitored and adjusted in real time to ensure that the candidate scheduling strategy obtained after each iteration does not cause the power system frequency to exceed the safe range, thereby avoiding potential frequency instability risks. After obtaining candidate solutions in each iteration of the alternating direction multiplier algorithm, the system can call the average system frequency ASF reduced-order model to evaluate the frequency impact of the system when subjected to disturbances, such as the current system frequency change rate, the current minimum point frequency deviation, and the current quasi-steady-state frequency deviation. The Average System Frequency (ASF) reduced-order model is a simplified dynamic model of power system frequency. By reducing computational complexity, it can predict the frequency change trend of the system in a short time. This model can be constructed based on parameters such as system inertia and load characteristics. If the calculation results show that any frequency index exceeds the preset safety threshold, it indicates that the current candidate scheduling strategy has the risk of frequency instability. At this time, the system will return a virtual inertia penalty term to the intra-park optimization subproblem in the alternating direction multiplier algorithm. This penalty term is proportional to the inertia deficit and can increase the optimization cost of the park under the condition of insufficient inertia. In the next iteration, it guides each park to actively adjust its heterogeneous resource scheduling scheme, so that frequency safety is verified and dynamically adjusted in real time at each step of the iterative solution. This ensures that the final generated cooperative scheduling strategy is safe while meeting the requirements of economy, thereby improving the overall reliability and safety of multi-park power system operation.

[0057] In one possible implementation, the power balance constraint can be:

[0058] Specifically, For the park exist The actual output of photovoltaic power at that time For the park exist The actual output of the photoelectric system at that time For the park exist The discharge power of the energy storage resources at that time. For the park exist The charging power of energy storage resources at that time. For the park exist The power exchanged between the power grid and the grid (positive for purchasing electricity, negative for selling electricity). For all other parks To the park The sum of the input mutual assistance power, For the park To all other parks The sum of the output mutual assistance power, For the park The baseline load, The sum of the regulation of all power resources in the park, such as impact load resources, continuous adjustable load resources and energy storage resources, can be obtained directly from the park by the source-grid-load-storage coordinated control system.

[0059] In one possible implementation, energy storage operation constraints may include:

[0060]

[0061]

[0062]

[0063] in, For the park State of charge of stored energy at t+1 For the park The state of charge of the stored energy at time t For the park The lower limit value corresponding to the energy storage state of charge. For the park The upper limit corresponding to the energy storage state of charge.

[0064] In one possible implementation, the tie-line capacity constraint can be: ,

[0065] in, For the park To the park Output mutual power, For the park and the park The upper limit of the mutual power between them.

[0066] In one possible implementation, resource-specific operational constraints may include: Used to constrain continuously adjustable loads; Used to constrain impulsive loads; interruption duration It is used to constrain the interruptible load of the computing center.

[0067] In one possible implementation, a multi-objective collaborative optimization model is constructed based on the resource adjustment model, including: constructing a total system operating cost function based on time-of-use electricity purchase price, time-of-use electricity sales price, the exchange power between each park and the target grid, and the adjustment cost of adjustable resources within each park; constructing a curtailment penalty cost function based on the curtailment penalty unit price and the amount of curtailment in each park; constructing a green revenue function based on the actual green electricity consumption and green electricity subsidy price; and constructing a multi-objective collaborative optimization model with the goal of minimizing the weighted sum by weighted summing of the total system operating cost function, the curtailment penalty cost function, and the green revenue function; wherein, the multi-objective collaborative optimization model is positively correlated with the total system operating cost function and the curtailment penalty cost function, and negatively correlated with the green revenue function.

[0068] Specifically, the multi-objective collaborative optimization model can be:

[0069] in, This represents the minimum total operating cost. This represents the total operating cost of the system. To cover the cost of power curtailment penalties, For green benefits; , , These are the first weight, the second weight, and the third weight, respectively.

[0070] Specifically, The corresponding total system operating cost function can be:

[0071] in, The electricity purchase price at time t; For the park The power consumption at time t; Let be the electricity price at time t. For the park The electricity sales power at time t; This refers to the cost coefficient for energy storage charging and discharging losses. For the park The charging and discharging power of energy storage resources; For power resources Adjustment costs; For power resources The adjustment amount, all of the above parameters can be obtained directly; The corresponding cost function for the power curtailment penalty can be:

[0072] in, The price is the penalty for abandoning electricity; For the park at time t The available photovoltaic power; For the park at time t The photovoltaic power absorption capacity, curtailment penalty price, available photovoltaic power, and photovoltaic power absorption capacity can all be obtained directly. The corresponding green revenue function can be:

[0073] in, Subsidizing the price of green electricity; For the park at time t The green electricity consumption capacity, green electricity subsidy price, and green electricity consumption capacity can all be obtained directly.

[0074] In one possible implementation, under safety constraints, the multi-objective collaborative optimization model is decomposed into an inter-park mutual assistance and grid exchange subproblem and an intra-park optimization subproblem using the alternating direction multiplier method. The solution is obtained by iteratively exchanging boundary information until convergence, yielding candidate scheduling strategies. These strategies include: in the inter-park mutual assistance and grid exchange subproblem, determining the adjustment amount of heterogeneous resources corresponding to each park as a fixed value and solving for it, to obtain the optimal mutual assistance power between parks and the optimal exchange power between each park and the grid when the sum of electricity purchase cost and mutual assistance cost is minimized; in the intra-park optimization subproblem, determining the optimal mutual assistance and grid exchange power between parks ...; in the intra-park optimization subproblem, determining the optimal mutual assistance and grid exchange power between parks and the grid; in the intra-park optimization subproblem, determining the optimal mutual assistance and grid exchange power between parks and the grid; in the intra-park optimization subproblem, determining In the sub-problem, the optimal mutual assistance power and the optimal exchange power are determined as fixed values. The local optimization objective is to minimize the adjustment cost within each park. The optimal adjustment amount of the heterogeneous resources corresponding to each park is solved separately. The solution is obtained by iteratively exchanging boundary information until convergence, and candidate scheduling strategies are obtained. In the iterative solution process, Lagrange multipliers are used to exchange boundary information between the inter-park mutual assistance and grid exchange sub-problems and the internal optimization sub-problems of each park. The boundary information includes the marginal price of mutual assistance power and the net load curve after adjustment within each park, until the convergence condition is met.

[0075] Specifically, the inter-park mutual assistance and grid exchange sub-problem is mainly used to handle power mutual assistance between multiple parks and interaction with the external power grid, in order to minimize the overall power purchase cost and mutual assistance cost. For example, the inter-park mutual assistance and grid exchange sub-problem can be:

[0076] in, To enhance the mutual support between this park and other parks, The electricity purchase price; The power consumption of the park; The electricity price; This refers to the electricity sales capacity of the industrial park. These are the preset penalty parameters; For the park in the first The mutual support power between the next iteration and other parks.

[0077] The adjustment amount corresponding to the heterogeneous resources in each park is ,in, For the park During the period The net load is the power that the park needs to be supplied by external sources (the power grid or other parks). A positive value indicates "power shortage" and a negative value indicates "surplus power that can be transmitted." For the park During the period The baseline load, i.e. the original power consumption before any adjustment, includes all production equipment, lighting, air conditioning and other conventional power consumption, which can be predicted based on historical power consumption data and user production plans; For the park During the period The actual output of photovoltaic power (or the photovoltaic power absorbed in the dispatch plan) can be predicted through numerical weather prediction (NWP) and photovoltaic power generation physical models, and will be affected by factors such as light intensity and temperature. For the park During the period The actual wind power output (or the wind power absorbed in the dispatch plan) can be predicted through numerical weather prediction (NWP) and wind power generation physical models, and is mainly affected by factors such as wind speed and wind direction. Based on numerical weather prediction and historical data, the system can generate wind and solar power output forecast curves (15-minute resolution) for each park for the next 24 hours. Based on historical electricity consumption data and user production plans for each park, baseline load forecast curves for each park for the next 24 hours are generated.

[0078] The optimization sub-problems within each park are used to adjust the operating status of various heterogeneous resources within the park to meet the power balance and operating constraints within the park. The optimization sub-problems within each park can be:

[0079] in, This represents the overall cost that the park needs to minimize in the current iteration. This is the unit operating cost coefficient (yuan / MWh) for energy storage systems, reflecting costs such as charging and discharging losses and depreciation, and can be obtained directly. This represents the energy storage charging and discharging power; positive indicates discharging, and negative indicates charging. For the first Adjustment cost function corresponding to adjustable load; The term is a Gram multiplier used to represent the marginal price (yuan / MWh) of mutual power exchange, which can guide the various parks to exchange power according to the principle of economic optimization. This represents the mutual support power between this park and other parks, with positive values ​​representing output and negative values ​​representing input. This is the frequency penalty coefficient, which controls the intensity of the penalty for violating frequency constraints (a positive constant). This is a penalty term for frequency constraint violation, which generates a positive value when a candidate solution in this park results in insufficient system inertia. The preset penalty parameter is used to control the cost of mutual assistance power deviating from the reference value in the optimization sub-problems within each park. Let be the reference mutual assistance power passed to each park for the inter-park mutual assistance and grid exchange subproblem, representing the mutual assistance value that the global coordinator expects the park to perform. Furthermore, in each iteration of the alternating direction multiplier method, the global coordinator updates the reference mutual assistance power as follows:

[0080] in, For each park in the first The actual output mutual aid power in the next iteration; For the updated Lagrange multipliers; The pre-set penalty parameters can be updated iteratively with the Lagrange multipliers, gradually guiding the independent optimization of each park to approach the global optimum.

[0081] During the iteration process, the inter-park mutual assistance and grid exchange subproblems update their Lagrange multipliers based on their calculation results and pass them on to the internal optimization subproblems of each park. Conversely, the internal optimization subproblems of each park also update their Lagrange multipliers based on their internal optimization results and pass them on to the inter-park mutual assistance and grid exchange subproblems. This allows each subproblem to be aware of the state and global constraints of other subproblems, thereby gradually adjusting its own decisions and ultimately achieving global optimum. The marginal price of mutual assistance power reflects the impact of increasing or decreasing unit mutual assistance power on the total system cost under the current scheduling state. The adjusted net load curve of each park refers to the actual load demand curve of each park after referring to the internal power resource adjustment, which can provide the inter-park mutual assistance optimization sub-model with the real power demand of each park. By decomposing the complex global optimization problem into several easily handled subproblems and using Lagrange multipliers as coordination variables to pass boundary information between subproblems, distributed collaborative optimization is achieved. This minimizes the sum of the total system operating cost, the cost of power curtailment penalty, and the benefit of green power consumption while ensuring the safe operation of the system.

[0082] This application discloses a method and system for collaborative scheduling of heterogeneous resources across multiple industrial parks for direct green power connections. This method is applied to a source-grid-load-storage collaborative control system, comprising: collecting real-time operating data and forecast data from multiple industrial parks; uniformly modeling different types of heterogeneous resources to obtain corresponding adjustment resource models for each resource; the heterogeneous resources include at least impact loads from coal mining areas, continuous adjustable loads from coal chemical industrial parks, interruptible loads from computing centers, and energy storage systems; all adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources; based on the adjustment resource models, a multi-objective collaborative optimization model and corresponding safety constraints are constructed; wherein, the multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits, while ensuring safety... The constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints transformed into linear inequalities through analytical processing. Under safety constraints, the alternating direction multiplier method is used to decompose the multi-objective collaborative optimization model into inter-park mutual assistance and grid exchange sub-problems, and intra-park optimization sub-problems. These sub-problems are solved iteratively by exchanging boundary information until convergence, yielding candidate scheduling strategies. Online frequency safety pre-evaluation and correction are performed on the candidate scheduling strategies. If the frequency index exceeds the safety threshold, a virtual inertia penalty term is returned to the intra-park optimization sub-problem, and the solution is iterated again until all frequency indexes meet the requirements, yielding the target collaborative scheduling strategy. Based on the target collaborative scheduling strategy, heterogeneous resources across multiple parks are collaboratively scheduled. This approach, by uniformly modeling heterogeneous power resources across multiple parks, constructing a multi-objective collaborative optimization model, and employing the alternating direction multiplier algorithm for distributed solution, can reduce the power operation costs of each park and improve the frequency safety and operational stability of the power system.

[0083] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a multi-park heterogeneous resource collaborative scheduling system, electronic equipment, and corresponding embodiments for direct green electricity connection.

[0084] Figure 2 This is a schematic diagram of the structure of a multi-park heterogeneous resource collaborative scheduling system for direct green electricity connection, as shown in an embodiment of this application.

[0085] See Figure 2 A multi-park heterogeneous resource collaborative scheduling system 200 for direct green electricity connection includes: The acquisition module 210 is used to collect real-time operation data and prediction data from multiple parks, and to uniformly model different types of heterogeneous resources to obtain the corresponding adjustment resource models for each resource. The heterogeneous resources include at least the impact load of the coal mining area, the continuous adjustable load of the coal chemical park, the interruptible load of the computing center, and the energy storage system. All adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources. Module 220 is used to construct a multi-objective collaborative optimization model and corresponding security constraints based on the resource adjustment model. The multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost and green benefits. The security constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints that are converted into linear inequalities through analytical processing. Processing module 230 is used to decompose the multi-objective collaborative optimization model into inter-park mutual assistance and power grid exchange sub-problems and internal optimization sub-problems of each park under safety constraints by using the alternating direction multiplier method. It solves the problem until convergence by iteratively exchanging boundary information to obtain candidate scheduling strategies. The correction module 240 is used to perform online pre-evaluation and correction of frequency security for candidate scheduling strategies. If the frequency index exceeds the security threshold, a virtual inertia penalty term is returned to the optimization sub-problem within the park, and the solution is iterated again until all frequency indexes meet the requirements, thus obtaining the target cooperative scheduling strategy. The collaborative scheduling module 250 is used to collaboratively schedule heterogeneous resources in multiple parks based on a target collaborative scheduling strategy.

[0086] In one possible implementation, the acquisition module 210 is further configured to model the impact load of the coal mining area based on the translation time window constraint and the total charging energy conservation constraint, to obtain a corresponding first regulating resource model; wherein the regulating amount of the first regulating resource model is the regulating amount of the impact load of the coal mining area relative to the reference power; model the continuous adjustable load of the coal chemical industrial park based on the power operating range constraint and the regulating rate constraint, to obtain a corresponding second regulating resource model; wherein the regulating amount of the second regulating resource model is the regulating amount of the continuous adjustable load of the coal chemical industrial park relative to the reference power; model the interruptible load of the computing center based on the interruption power constraint and the interruption duration constraint, to obtain a corresponding third regulating resource model; wherein the regulating amount of the third regulating resource model is the regulating amount of the interruption power corresponding to the interruptible load of the computing center; and model the energy storage system based on the charging and discharging power range constraint and the state of charge dynamic constraint, to obtain a corresponding fourth regulating resource model; wherein the regulating amount of the fourth regulating resource model is the regulating amount of the charging and discharging power of the energy storage system.

[0087] In one possible implementation, the construction module 220 further calculates the virtual inertia time constant of each park in each time period based on the ratio of grid-type equipment to grid-type equipment in each park, the operating status of the energy storage system, and the power of the tie line; sums the virtual inertia time constants of each park to obtain the total system inertia of the source-grid-load-storage collaborative control system in each time period; and constructs linear inequality constraints related to the total system inertia based on the total system inertia, the regulation resource model, the system frequency change rate, and the minimum frequency deviation.

[0088] In one possible implementation, the construction module 220 is further configured to obtain the power disturbance amount when the source-grid-load-storage coordinated control system experiences a preset maximum power disturbance; based on the power disturbance amount, system damping ratio, natural oscillation frequency, and total system inertia, obtain a first expression corresponding to the system frequency change rate of the source-grid-load-storage coordinated control system and a second expression corresponding to the minimum frequency deviation; wherein, the system frequency change rate is less than a first safety threshold, and the minimum frequency deviation is less than a second safety threshold; and construct linear inequality constraints related to the total system inertia based on the first and second expressions.

[0089] In one possible implementation, the correction module 240 is further configured to input the current total system inertia and the current total regulation power corresponding to the current candidate scheduling strategy into the average system frequency reduction model to obtain the current system frequency change rate, the current minimum point frequency deviation, and the current quasi-steady-state frequency deviation corresponding to the current candidate scheduling strategy; if the current system frequency change rate exceeds the first safety threshold, and / or the current minimum point frequency deviation exceeds the second safety threshold, and / or the current quasi-steady-state frequency deviation exceeds the third safety threshold, then a virtual inertia penalty term is generated; the virtual inertia penalty term is returned to the internal optimization sub-problem of the park, and iteratively solved again until all frequency indicators meet the requirements to obtain the target cooperative scheduling strategy.

[0090] In one possible implementation, the construction module 220 is further configured to construct a total system operating cost function based on the time-of-use electricity purchase price, the time-of-use electricity sales price, the exchange power between each park and the target grid, and the adjustment cost of adjustable resources within each park; construct a curtailment penalty cost function based on the curtailment penalty unit price and the amount of curtailment in each park; construct a green revenue function based on the actual green electricity consumption and the green electricity subsidy price; and perform a weighted summation of the total system operating cost function, the curtailment penalty cost function, and the green revenue function to construct a multi-objective collaborative optimization model with the goal of minimizing the weighted sum; wherein, the multi-objective collaborative optimization model is positively correlated with the total system operating cost function and the curtailment penalty cost function, and negatively correlated with the green revenue function.

[0091] In one possible implementation, the processing module 230 is further configured to, in the inter-park mutual assistance and grid exchange subproblem, determine the adjustment amount of the heterogeneous resources corresponding to each park as a fixed value and solve it to obtain the optimal mutual assistance power between each park and the optimal exchange power between each park and the grid when the sum of the power purchase cost and mutual assistance cost is minimized; in the optimization subproblem within each park, the optimal mutual assistance power and the optimal exchange power are determined as fixed values, and the local optimization objective is to minimize the adjustment cost within each park. The optimal adjustment amount of the heterogeneous resources corresponding to each park is solved separately, and the solution is obtained by iteratively exchanging boundary information until convergence, thus obtaining a candidate scheduling strategy; wherein, in the iterative solution process, boundary information is exchanged between the inter-park mutual assistance and grid exchange subproblem and the optimization subproblem within each park through Lagrange multipliers. The boundary information includes the marginal price of mutual assistance power and the net load curve after adjustment within each park, until the convergence condition is met.

[0092] This application discloses a multi-park heterogeneous resource collaborative scheduling system for direct green power connection, comprising: an acquisition module for collecting real-time operation data and forecast data from multiple parks, uniformly modeling different types of heterogeneous resources, and obtaining adjustment resource models corresponding to each resource; the heterogeneous resources include at least the impact load of coal mining areas, the continuously adjustable load of coal chemical industrial parks, the interruptible load of computing centers, and energy storage systems, and all adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of heterogeneous resources; and a construction module for constructing a multi-objective collaborative optimization model and corresponding security constraints based on the adjustment resource models; wherein, the multi-objective collaborative optimization model takes the minimization of the weighted sum of the total system operating cost, the cost of power curtailment penalty, the cost of carbon emissions, and the green benefits as the optimization objective, and the security constraints include at least power balance constraints, The system comprises several modules: an energy storage operation constraint, tie-line capacity constraint, resource-specific operation constraint, and frequency stability constraint transformed into a linear inequality through analytical processing; a processing module, under safety constraints, decomposes the multi-objective collaborative optimization model into inter-park mutual assistance and grid exchange sub-problems and intra-park optimization sub-problems using the alternating direction multiplier method; a correction module performs online pre-evaluation and correction of the candidate scheduling strategies for frequency safety; if the frequency index exceeds the safety threshold, a virtual inertia penalty term is returned to the intra-park optimization sub-problem, and the solution is iterated again until all frequency indexes meet the requirements, yielding the target collaborative scheduling strategy; and a collaborative scheduling module performs collaborative scheduling of heterogeneous resources across multiple parks based on the target collaborative scheduling strategy. This approach, by uniformly modeling heterogeneous power resources across multiple parks, constructing a multi-objective collaborative optimization model, and employing the alternating direction multiplier algorithm for distributed solution, can reduce the power operation costs of each park and improve the frequency safety and operational stability of the power system.

[0093] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0094] This application also provides an electronic device. Figure 3 This is a schematic diagram of the hardware structure of an embodiment of the electronic device of this application. The electronic device includes a memory 320 and at least one processor 310. The memory 320 is electrically connected to the at least one processor 310. The memory 320 stores instructions. The at least one processor 310 calls the instructions in the memory 320, causing the electronic device to execute the multi-campus heterogeneous resource collaborative scheduling method for direct green power connection according to any of the foregoing embodiments of this application.

[0095] Specifically, the processor 310 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0096] Memory 320 may include a mass storage device for data or instructions. For example, and not limitingly, memory 320 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 320 may include removable or non-removable (or fixed) media. Where appropriate, memory 320 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 320 is non-volatile solid-state memory. In a particular embodiment, memory 320 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0097] In one example, the control device may also include a communication interface 330 and a bus 340. The processor 310, memory 320, and communication interface 330 are connected via the bus 340 and communicate with each other.

[0098] The communication interface 330 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0099] Bus 340 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 340 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0100] Furthermore, in conjunction with the multi-campus heterogeneous resource collaborative scheduling method for direct green power connection described in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores executable code, which, when executed by a processor, implements any of the multi-campus heterogeneous resource collaborative scheduling methods for direct green power connection described in the above embodiments.

[0101] This application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0102] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0103] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0104] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for collaborative scheduling of heterogeneous resources in multiple parks for direct green electricity connection, characterized in that, Applications include: Source-grid-load-storage coordinated control systems Real-time operation data and forecast data from multiple parks are collected, and different types of heterogeneous resources are uniformly modeled to obtain the corresponding adjustment resource models for each resource. The heterogeneous resources include at least the shock load of coal mining areas, the continuous adjustable load of coal chemical industrial parks, the interruptible load of computing centers, and energy storage systems. All the adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources. Based on the aforementioned resource adjustment model, a multi-objective collaborative optimization model and corresponding security constraints are constructed. The multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits. The security constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints that are transformed into linear inequalities through analytical processing. Under the aforementioned safety constraints, the multi-objective collaborative optimization model is decomposed into inter-park mutual assistance and power grid exchange sub-problems and internal optimization sub-problems within each park using the alternating direction multiplier method. The solution is obtained by iteratively exchanging boundary information until convergence, thus yielding candidate scheduling strategies. The candidate scheduling strategy is subjected to online pre-evaluation and correction for frequency security. If the frequency index exceeds the security threshold, the virtual inertia penalty term is returned to the optimization sub-problem within the park, and the solution is iterated again until all frequency indexes meet the requirements, thus obtaining the target cooperative scheduling strategy. Based on the aforementioned target collaborative scheduling strategy, heterogeneous resources from multiple parks are collaboratively scheduled.

2. The method according to claim 1, characterized in that, The unified modeling of different types of heterogeneous resources to obtain the corresponding adjustment resource model for each resource includes: Based on the constraints of translation time window and total charging energy conservation, the impact load of the coal mining area is modeled to obtain the corresponding first adjustment resource model; wherein, the adjustment amount of the first adjustment resource model is the adjustment amount of the impact load of the coal mining area relative to the reference power. Based on power operating range constraints and adjustment rate constraints, the continuous adjustable load of the coal chemical industrial park is modeled to obtain the corresponding second adjustment resource model; wherein, the adjustment amount of the second adjustment resource model is the adjustment amount of the continuous adjustable load of the coal chemical industrial park relative to the reference power. Based on interruption power constraints and interruption duration constraints, the interruptible load of the computing center is modeled to obtain the corresponding third adjustment resource model; wherein, the adjustment amount of the third adjustment resource model is the adjustment amount of the interruption power corresponding to the interruptible load of the computing center. Based on the charging and discharging power range constraints and the dynamic state of charge constraints, the energy storage system is modeled to obtain the corresponding fourth regulation resource model; wherein, the regulation amount of the fourth regulation resource model is the regulation amount of the charging and discharging power of the energy storage system.

3. The method according to claim 1, characterized in that, The construction of frequency stability constraints based on the aforementioned resource adjustment model, which are transformed into linear inequalities through analytical processing, includes: Based on the ratio of internal grid-type equipment to grid-type equipment in each park, the operating status of the energy storage system, and the power of the tie line, the virtual inertial time constant of each park in each time period is calculated. The total system inertia of the source-grid-load-storage coordinated control system in each time period is obtained by summing the virtual inertial time constants of each park. Based on the total system inertia, the adjustment resource model, the system frequency change rate, and the minimum frequency deviation, a linear inequality constraint related to the total system inertia is constructed.

4. The method according to claim 3, characterized in that, The step of constructing linear inequality constraints related to the total inertia of the system based on the total system inertia, the adjustment resource model, the maximum power change value, and the minimum frequency deviation value includes: Obtain the power disturbance amount when the source-grid-load-storage coordinated control system experiences a preset maximum power disturbance; Based on the power disturbance, system damping ratio, natural oscillation frequency, and total system inertia, a first expression for the system frequency change rate and a second expression for the minimum frequency deviation of the source-grid-load-storage coordinated control system are obtained; wherein, the system frequency change rate is less than a first safety threshold, and the minimum frequency deviation is less than a second safety threshold. Based on the first expression and the second expression, construct linear inequality constraints related to the total inertia of the system.

5. The method according to claim 4, characterized in that, The process involves online pre-evaluation and correction of the candidate scheduling strategy for frequency security. If the frequency index exceeds the security threshold, a virtual inertia penalty term is returned to the internal optimization sub-problem of the park, and the solution is iterated again until all frequency indices meet the requirements, thus obtaining the target cooperative scheduling strategy, including: Input the current system total inertia and current total regulation power corresponding to the current candidate scheduling strategy into the average system frequency reduction model to obtain the current system frequency change rate, current minimum point frequency deviation and current quasi-steady state frequency deviation corresponding to the current candidate scheduling strategy; If the current system frequency change rate exceeds the first safety threshold, and / or the current minimum point frequency deviation exceeds the second safety threshold, and / or the current quasi-steady-state frequency deviation exceeds the third safety threshold, then the virtual inertia penalty term is generated; The virtual inertia penalty term is returned to the internal optimization subproblem of the park and iterated again until all frequency indicators meet the requirements, thus obtaining the target cooperative scheduling strategy.

6. The method according to claim 1, characterized in that, The construction of a multi-objective collaborative optimization model based on the resource adjustment model includes: Based on the time-of-use electricity purchase price, time-of-use electricity sales price, the exchange power between each park and the target power grid, and the adjustment cost of adjustable resources within each park, a total system operating cost function is constructed. Based on the unit price of power curtailment penalty and the amount of power curtailment in each park, a power curtailment penalty cost function is constructed; Construct a green revenue function based on the actual amount of green electricity consumed and the price of green electricity subsidies; The total system operating cost function, the curtailment penalty cost function, and the green revenue function are weighted and summed to construct a multi-objective collaborative optimization model with the goal of minimizing the weighted sum; wherein the multi-objective collaborative optimization model is positively correlated with the total system operating cost function and the curtailment penalty cost function, and negatively correlated with the green revenue function.

7. The method according to claim 1, characterized in that, Under the aforementioned safety constraints, the multi-objective collaborative optimization model is decomposed into inter-park mutual assistance and grid exchange sub-problems and intra-park optimization sub-problems using the alternating direction multiplier method. The model is then solved iteratively by exchanging boundary information until convergence, yielding candidate scheduling strategies, including: In the subproblem of mutual assistance and grid exchange between the parks, the adjustment amount of the heterogeneous resources corresponding to each park is determined as a fixed value and solved to obtain the optimal mutual assistance power between the parks and the optimal exchange power between each park and the grid when the sum of the power purchase cost and mutual assistance cost is minimized. In the optimization sub-problems within each park, the optimal mutual assistance power and the optimal exchange power are determined as fixed values. The local optimization objective is to minimize the adjustment cost within each park. The optimal adjustment amount of the heterogeneous resources corresponding to each park is solved respectively. The solution is obtained by iteratively exchanging boundary information until convergence, thus obtaining the candidate scheduling strategy. In the iterative solution process, Lagrange multipliers are used to exchange boundary information between the inter-park mutual assistance and grid exchange subproblems and the internal optimization subproblems of each park. The boundary information includes the marginal price of mutual assistance power and the adjusted net load curve of each park, until the convergence condition is met.

8. A multi-park heterogeneous resource collaborative scheduling system for direct green electricity connection, characterized in that, include: The acquisition module is used to collect real-time operation data and prediction data from multiple parks, and to uniformly model different types of heterogeneous resources to obtain the corresponding adjustment resource models for each resource. The heterogeneous resources include at least the impact load of coal mining areas, the continuous adjustable load of coal chemical industrial parks, the interruptible load of computing centers, and energy storage systems. All the adjustment resource models are uniformly associated with the adjustment amount and adjustment cost of the heterogeneous resources. The construction module is used to construct a multi-objective collaborative optimization model and corresponding security constraints based on the aforementioned resource adjustment model. The multi-objective collaborative optimization model aims to minimize the weighted sum of the total system operating cost, curtailment penalty cost, carbon emission cost, and green benefits. The security constraints include at least power balance constraints, energy storage operation constraints, tie-line capacity constraints, resource-specific operation constraints, and frequency stability constraints that are converted into linear inequalities through analytical processing. The processing module is used to decompose the multi-objective collaborative optimization model into inter-park mutual assistance and power grid exchange sub-problems and internal optimization sub-problems within each park under the security constraints using the alternating direction multiplier method, and solve the problem until convergence by iteratively exchanging boundary information to obtain candidate scheduling strategies. The correction module is used to perform online pre-evaluation and correction of the frequency security of the candidate scheduling strategy. If the frequency index exceeds the security threshold, the virtual inertia penalty term is returned to the optimization sub-problem within the park, and the solution is iterated again until all frequency indexes meet the requirements, and the target cooperative scheduling strategy is obtained. The collaborative scheduling module is used to collaboratively schedule heterogeneous resources in multiple parks based on the target collaborative scheduling strategy.

9. An electronic device, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores executable code that, when executed by a processor of an electronic device, causes the processor to perform the method as described in any one of claims 1-7.

Images