Cross-regional new energy power system coordination optimization method and device

Abstract

The invention provides a cross-regional new energy power system coordinated optimization method and device, and relates to the technical field of power system operation and control, and the method comprises the steps: constructing a two-stage distributed optimization model, enabling the optimization target of an upper coordinator to minimize the standby resource cost of the system and the transmission loss of a cross-regional tie line, and enabling the optimization target of a lower coordinator to be the minimum standby resource cost of the system; the optimization target of the area controller is to minimize the operation cost of the area and the risk loss based on the condition risk value; in the rolling time domain of the control period, collaborative iterative calculation is executed through an alternating direction multiplier method; and when a preset convergence condition is met, the region executes a region internal scheduling scheme at the current moment, and forwards rolls a time sequence by a preset step length to perform collaborative iterative calculation of a next control period. According to the method, data privacy and efficient calculation are guaranteed, collaborative optimization of cross-regional power and standby resources is achieved, and the economical efficiency, safety and real-time scheduling capacity of a high-proportion new energy power system for coping with uncertain risks are improved.

Description

Technical Field

[0001] This invention relates to the field of power system operation and control technology, and in particular to a method and apparatus for coordinated optimization of cross-regional new energy power systems. Background Technology

[0002] With the large-scale grid connection of renewable energy sources such as wind power and photovoltaics, the power system is accelerating its evolution towards a new architecture with a high proportion of new energy and cross-regional interconnection. Based on this, the implementation of cross-regional power transmission projects such as the West-to-East Power Transmission and the North-to-South Power Transmission has made the operation of the power system exhibit complex characteristics of multi-level and wide-area coordination. However, considering the inherent strong randomness, volatility and intermittency of new energy output, coupled with the spatial mismatch between its resource-rich areas and load centers, the cross-regional power exchange plan is frequently adjusted and the power flow direction is constantly changing, posing unprecedented challenges to the real-time performance, coordination and risk response capabilities of dispatching.

[0003] Current mainstream cross-regional scheduling methods mainly fall into two categories: centralized optimization and decentralized autonomous methods. While centralized optimization methods can achieve global optimization, they require the collection of detailed data from the entire network, resulting in problems such as heavy communication burden, high computational complexity, and significant privacy leakage risks, making it difficult to meet the timeliness requirements of real-time scheduling at the 5 to 15 minute level. On the other hand, while decentralized autonomous methods protect regional independence, they lack an effective global coordination mechanism, which can easily lead to inconsistent tie-line power, unbalanced reserve allocation, and power flow oscillations. Furthermore, they cannot quantify and collaboratively suppress the tail risks brought about by the uncertainty of new energy sources. In addition, existing stochastic or robust optimization methods have low computational efficiency when dealing with high-dimensional scenarios, making it difficult to match the rolling scheduling cycle.

[0004] Therefore, a method for coordinating and optimizing cross-regional renewable energy power systems, which can achieve rapid coordination and allocation of cross-regional power and reserve resources while ensuring data privacy and computational scalability, and can effectively quantify and suppress extreme risks of renewable energy fluctuations, urgently needs to be studied. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention provides a method and apparatus for coordinated optimization of cross-regional new energy power systems. This invention solves the technical problems in the existing technologies, such as the difficulty of centralized optimization in balancing real-time performance and privacy security, the inability of decentralized autonomous methods to solve the problems of boundary power consistency and collaborative suppression of system risks, and the low computational efficiency of existing optimization algorithms in high-dimensional uncertainty scenarios.

[0006] This invention provides a method for coordinated optimization of cross-regional new energy power systems, comprising: A two-level distributed optimization model is constructed, wherein the two-level distributed optimization model includes an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's backup resource cost and cross-area tie line transmission loss. The optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk value. Within the rolling time domain of each control cycle, the upper-level coordinator performs collaborative iterative calculations between the multiple area controllers using the alternating direction multiplier method. The collaborative iterative calculations include area update calculations, coordinator update calculations, and multiplier update calculations. When the collaborative iterative calculation meets the preset convergence condition, the region executes the regional internal scheduling scheme determined by the collaborative iterative calculation for the current moment, and rolls the time sequence forward by a preset step, and performs collaborative iterative calculation for the next control cycle based on the updated system data.

[0007] Optionally, the constraints of the optimization model of the upper-level coordinator include at least the following: the sum of the power exchanged between the region and all adjacent regions is equal to the net injected power of the region; the transmission power of the tie line is limited by capacity, and the power change between adjacent time periods is limited by ramp rate; the power exchanged between any two regions satisfies a reciprocal relationship, and the sum of the net injected power of the system is zero; the sum of the upper and lower reserve capacities provided by all regions is not less than the total reserve requirement of the system.

[0008] Optionally, the constraints of the optimization function of the regional controller include at least the following: the sum of the power generation of all conventional units and available renewable energy in the region meets the needs of regional load and external interconnection power; the output of conventional units is within the preset range of technical output, and the output change in adjacent time periods is limited by ramping capacity; the available output of renewable energy is decomposed into actual utilization power and planned curtailment power, and the planned curtailment power is a non-negative value.

[0009] Optionally, the conditional risk value is calculated in a linear form, and the calculation process depends on a quantile variable and a set of non-negative relaxation variables corresponding to the new energy output scenarios; wherein, the conditional risk value is expressed as the sum of the quantile value and an adjustment term, the adjustment term is calculated based on the relaxation variables under different scenarios according to a preset confidence level, and the relaxation variable under each scenario is constrained to be no less than the difference between the operating loss under the corresponding scenario and the quantile value, while maintaining non-negativity.

[0010] Optionally, the region update calculation includes: the region controller, based on the current reference power and shadow price issued by the upper-level coordinator, solves and updates the boundary power plan with the goal of minimizing the sum of the local optimization objective and the consistency penalty term; the coordinator update calculation includes: the upper-level coordinator aggregates all the updated boundary power plans of the region controllers, solves and updates the reference power sent to the region with the goal of minimizing the sum of the global optimization objective and the consistency penalty term; the multiplier update calculation includes: the upper-level coordinator updates the shadow price sent to the region based on a preset rule according to the deviation between the updated boundary power plan of the region and its own updated reference power.

[0011] Optionally, the preset convergence conditions include: the original residual between the boundary power plan and the reference power of the region is lower than a first residual threshold, and the dual residual of the reference power change between adjacent iterations is lower than a second residual threshold.

[0012] Optionally, the regional controller operates based on a rolling predictive control mechanism, which includes: at the beginning of each control cycle, calculating the optimization function of the regional controller based on the load and renewable energy output forecast data within the forecast window starting from the current time; after obtaining the control instructions within the forecast window, implementing the first control instruction corresponding to the current time; rolling the forecast window forward by a preset step, and repeating the optimization and execution process in the next control cycle based on the updated system state and forecast data, while receiving the updated price signal from the upper-level coordinator, forming a continuous closed-loop optimization control.

[0013] Optionally, the penalty parameter in the alternating direction multiplier method may be a fixed value or dynamically adjusted according to the relative magnitude of the original residual and the dual residual.

[0014] Optionally, when the area includes an energy storage unit, the constraints of the area controller include at least the following: the energy storage state of the energy storage unit is dynamically updated based on the charging power, discharging power and corresponding charging and discharging efficiency; the energy storage state of the energy storage unit is limited to between zero and the rated capacity; and the charging power and discharging power of the energy storage unit are respectively limited to between zero and the maximum power limit.

[0015] Another aspect of the present invention provides a cross-regional new energy power system coordination and optimization device, comprising: The model building unit is used to build a two-level distributed optimization model, wherein the two-level distributed optimization model includes an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's backup resource cost and cross-area tie-line transmission loss. The optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk value. The collaborative iteration unit is used to perform collaborative iteration calculations between the upper-level coordinator and multiple region controllers in the rolling time domain of each control cycle by using the alternating direction multiplier method. The collaborative iteration calculations include region update calculations, coordinator update calculations, and multiplier update calculations. The execution update unit is used to execute the regional internal scheduling scheme determined by the collaborative iterative calculation for the current time when the collaborative iterative calculation meets the preset convergence condition, and to roll the time sequence forward by a preset step size, and to perform collaborative iterative calculation for the next control cycle based on the updated system data.

[0016] The present invention provides a method and apparatus for coordinated optimization of cross-regional new energy power systems. First, by constructing a two-level distributed optimization model consisting of an upper-level coordinator and multiple regional controllers, a coordination mechanism is introduced at the upper level while retaining the autonomous decision-making capabilities of each region. This effectively avoids the communication bottlenecks, high computational complexity, and privacy leakage risks associated with centralized methods that require data collection from the entire network, thus meeting the real-time requirements of rolling scheduling. Second, the upper-level coordinator aims to minimize the cost of reserve resources and tie-line losses, while the regional controllers consider both operating costs and risk losses based on conditional risk values, achieving a balance between economy and security. This enables the system to maintain high utilization of renewable energy while possessing stronger anti-disturbance capabilities. Finally, the alternating direction multiplier method is used to perform collaborative iterative calculations in the rolling time domain, and the scheduling scheme is updated rollingly after convergence. This ensures the consistency and global coordination of solutions across multiple regions and also possesses good computational scalability, improving the efficiency and robustness of cross-regional power and reserve resource coordination and allocation in high-dimensional stochastic scenarios. The above method, while ensuring data privacy and computational efficiency, enables rapid collaborative optimization of cross-regional power and reserve resources, and improves the economy, security and real-time dispatch capabilities of high-proportion renewable energy power systems in response to uncertainty risks.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1A schematic diagram of the overall process of the cross-regional new energy power system coordination and optimization method provided in one embodiment of this application; Figure 2 A schematic diagram of the overall architecture of a two-level distributed optimization model in a cross-regional new energy power system coordination optimization method provided in this application; Figure 3 A schematic diagram of the rolling model predictive control mechanism in the cross-regional new energy power system coordination optimization method provided in one embodiment of this application; Figure 4 A schematic diagram of the structure of a cross-regional new energy power system coordination and optimization device provided in one embodiment of this application. Detailed Implementation

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0021] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0022] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] This invention provides a method for coordinated optimization of cross-regional new energy power systems, such as... Figure 1As shown, the process includes: constructing a two-level distributed optimization model, which comprises an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's reserve resource cost and cross-area tie-line transmission loss, while the optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk values. Within the rolling time domain of each control cycle, a collaborative iterative calculation is performed between the upper-level coordinator and multiple area controllers using the alternating direction multiplier method. This collaborative iterative calculation includes area update calculation, coordinator update calculation, and multiplier update calculation. When the collaborative iterative calculation meets the preset convergence condition, the area executes the intra-area scheduling scheme determined by the collaborative iterative calculation for the current moment, and rolls the time sequence forward by a preset step. Based on the updated system data, the collaborative iterative calculation for the next control cycle is then performed.

[0024] The cross-regional renewable energy power system coordination optimization method provided by this invention firstly constructs a two-level distributed optimization model consisting of an upper-level coordinator and multiple regional controllers. While preserving the autonomous decision-making capabilities of each region, it introduces an upper-level coordination mechanism, effectively avoiding the communication bottlenecks, high computational complexity, and privacy leakage risks associated with centralized methods that require data aggregation across the entire network, thus meeting the real-time requirements of rolling scheduling. Secondly, the upper-level coordinator aims to minimize the cost of reserve resources and tie-line losses, while the regional controllers balance operating costs and risk losses based on conditional risk values, achieving a balance between economy and security. This enables the system to maintain high utilization of renewable energy while possessing stronger anti-disturbance capabilities. Finally, the alternating direction multiplier method is used to perform collaborative iterative calculations in the rolling time domain, and the scheduling scheme is updated rollingly after convergence. This ensures the consistency and global coordination of solutions across multiple regions, while also possessing good computational scalability, improving the efficiency and robustness of cross-regional power and reserve resource coordination allocation in high-dimensional stochastic scenarios. The above method, while ensuring data privacy and computational efficiency, enables rapid collaborative optimization of cross-regional power and reserve resources, and improves the economy, security and real-time dispatch capabilities of high-proportion renewable energy power systems in response to uncertainty risks.

[0025] Specifically, in the above embodiments, the constraints of the optimization model of the upper coordinator include at least the following: the sum of the power exchanged between the region and all adjacent regions is equal to the net injected power of the region; the transmission power of the tie line is limited by capacity, and the power change between adjacent time periods is limited by ramp rate; the power exchanged between any two regions satisfies a reciprocal relationship, and the sum of the net injected power of the system is zero; the sum of the upper and lower reserve capacity provided by all regions is not less than the total reserve requirement of the system.

[0026] In this embodiment, the optimization objective of the upper-layer coordinator is to minimize the system's backup resource cost and cross-regional tie-line transmission loss. The optimization model is as follows:

[0027] In the formula, T represents the prediction time domain; Let be the power of the tie line from region r to the adjacent region s; , These are the upper and lower reserve power outputs for region r, respectively. For system backup costs; Cost of transmission loss for the tie line; The specific constraints of the optimization model include: regional net injected power balance.

[0028] Tie line capacity and power change rate limits:

[0029]

[0030] System power balance and reciprocal relationship:

[0031]

[0032] System backup adequacy:

[0033]

[0034] In the formula, Let r be the net injected power in region r; The set of adjacent regions connected to region r; , These are the tie line power capacity and the ramp limit for adjacent times, respectively; , These are the system's upper backup requirements and lower backup requirements, respectively.

[0035] Specifically, in the above embodiments, the constraints of the optimization function of the area controller include at least: The sum of the power generation of all conventional generating units and available renewable energy sources in the region meets the needs of regional load and external interconnection power; the output of conventional generating units is within the preset range of technical output, and the output variation between adjacent time periods is limited by ramping capacity; the available output of renewable energy is decomposed into actual utilization power and planned curtailment power, and the planned curtailment power is a non-negative value.

[0036] In this embodiment, the optimization objective of the area controller is to minimize the operating cost of the area and the risk loss based on the conditional risk value. The optimization function is:

[0037] In the formula, H represents the current control time; H is the prediction window length. Fuel costs for conventional generating units; Punishment for abandoning wind and light; The shadow price returned by the upper-level coordinator; Let r be the net injected power in region r; This refers to the risk weighting coefficient. Confidence level Conditional risk value; The specific constraints of the optimization function include: Regional balance:

[0038] Unit output and ramp-up limitations:

[0039]

[0040] Renewable power decomposition:

[0041]

[0042] In the formula: For the unit g output; and These are the minimum power generation capacity and the maximum power generation capacity, respectively. This refers to the amount of electricity wasted. It is renewable power; For the load of region r; 0 represents the predicted available renewable power output; 0 represents the unit ramp rate.

[0043] Specifically, in the above embodiments, the conditional risk value is calculated in a linearized form, and the calculation process depends on a quantile variable and a set of non-negative relaxation variables corresponding to the new energy output scenario. The conditional risk value is represented as the sum of the quantile and an adjustment term. The adjustment term is calculated based on the slack variables under different scenarios according to a preset confidence level. The slack variables under each scenario are constrained to be no less than the difference between the running loss and the quantile under the corresponding scenario, while remaining non-negative.

[0044] In this embodiment, the conditional risk value CVaR is linearized to improve the solvability of the model, and the specific expression is as follows:

[0045]

[0046] In the formula: A collection of scenarios contributing to new energy; Number of scenes; This refers to the quantile value (i.e., the risk threshold). For the scene Excess loss slack variables; For the scene The running loss function.

[0047] Specifically, in the above embodiments, the region update calculation includes: the region controller, based on the current reference power and shadow price issued by the upper-layer coordinator, solves and updates the boundary power plan with the goal of minimizing the sum of the local optimization objective and the consistency penalty term; the coordinator update calculation includes: the upper-layer coordinator aggregates the updated boundary power plans of all region controllers, solves and updates the reference power sent to the region with the goal of minimizing the sum of the global optimization objective and the consistency penalty term; the multiplier update calculation includes: the upper-layer coordinator updates the shadow price sent to the region based on a preset rule according to the deviation between the updated boundary power plan of the region and its own updated reference power.

[0048] Specifically, in the above embodiments, the preset convergence conditions include: The original residual between the boundary power plan and the reference power of the region is lower than the first residual threshold, and the dual residual of the reference power change between adjacent iterations is lower than the second residual threshold.

[0049] In this embodiment, the region update calculation formula is:

[0050] The coordinator update calculation formula is:

[0051] The formula for multiplier update is:

[0052] And set the convergence condition as follows:

[0053]

[0054] In the formula: k is the number of iterations; The reference power for the upper-level coordinator; ρ is a Lagrange multiplier (shadow price); ρ>0 is a penalty parameter; The original residual threshold, This is the threshold for the dual residual.

[0055] Specifically, in the above embodiments, the regional controller operates based on a rolling predictive control mechanism, which includes: at the beginning of each control cycle, calculating the regional controller's optimization function based on the load and renewable energy output forecast data within the forecast window starting from the current time; after obtaining the control instructions within the forecast window, implementing the first control instruction corresponding to the current time; rolling the forecast window forward by a preset step, and repeating the optimization and execution process in the next control cycle based on the updated system state and forecast data, while receiving updated price signals from the upper-level coordinator, forming a continuous closed-loop optimization control.

[0056] In this embodiment, the area controller employs a rolling predictive control mechanism, at each control time... Collection interval [ , +H The load and renewable energy forecast data within [1] are used to solve the equation and execute only the first step of control. Subsequently, the forecast information and upper-level price signals are updated continuously to form a continuous closed-loop optimization.

[0057] Specifically, in the above embodiments, when the region includes an energy storage unit, the constraints of the region controller include at least the following: the energy storage state of the energy storage unit is dynamically updated according to the charging power, discharging power and the corresponding charging and discharging efficiency; the energy storage state of the energy storage unit is limited to between zero and the rated capacity; and the charging power and discharging power of the energy storage unit are respectively limited to between zero and the maximum power limit.

[0058] In this embodiment, when the region includes energy storage units, the energy storage energy balance equation is:

[0059] And satisfy the constraints:

[0060]

[0061]

[0062] In the formula, It is a stored energy state; , These are the charge and discharge efficiencies, respectively. , These are the charging and discharging powers, respectively. , , These are the upper limits for energy and power, respectively.

[0063] Specifically, in the above embodiments, the penalty parameter in the alternating direction multiplier method is a fixed value, or is dynamically adjusted according to the relative magnitude of the original residual and the dual residual.

[0064] In this embodiment, the time step, prediction window length, number of scenarios, and confidence level can be adaptively set according to the system scale and real-time requirements. The time step is 5 to 15 minutes, the prediction window is 1 to 4 hours, the number of scenarios is 10 to 100, and the confidence level α is 0.90–0.99. The penalty parameter ρ can adopt a fixed value or an adaptive adjustment strategy to accelerate convergence. The specific form of each cost function can be a linear or quadratic function, as follows:

[0065]

[0066]

[0067]

[0068] In the formula: For the unit's fuel cost, As a penalty for abandoning electricity, For backup costs at the upper level, For backup costs at lower levels, This represents the power flow loss coefficient.

[0069] Based on the cross-regional new energy power system coordination and optimization method provided in this application, the specific execution process is as follows: The system execution process is as follows: Figure 2 As shown, the process includes: initialization and data preparation, where each regional controller maintains model parameters for units, networks, energy storage, and new energy sources locally, while the upper-level coordinator maintains global parameters such as inter-regional tie lines and system-wide reserve requirements (without exchanging internal private data); upper-level reference and signal distribution, at the beginning of each rolling period, the upper layer forms a reference for tie line power and reserve based on the previous period's results and the latest boundary information, and distributes price / shadow price signals to guide each region; regional local rolling optimization, where each regional controller performs window optimization based on local constraints and predictions, generates local scheduling and boundary injection power, and only returns boundary-related quantities and target values, without returning private parameters; and coordination and updating, where the upper layer aggregates boundary power and cost information for each region, updates the reference and price signals for the next round, until consistency reaches the threshold.

[0070] Rolling model predictive control mechanism such as Figure 3 As shown, this includes: collecting predictions at the current control moment. Collect prediction windows Load and renewable energy output prediction within the region; solution window, rolling optimization within the solution region to obtain the results from... arrive The control sequence; execute the first step, only implementing... The control parameters are set, and other steps are reserved as references; the right-moving window and update are constantly advancing to the desired state. +1, move the window to the right. And re-optimize using the latest forecasts; at the same time, apply the new coordinated price signals from the upper level to ensure consistency with... Figure 2 Cross-regional coordination and connection.

[0071] The implementation and effects of the example include: Implementation verification was conducted on the IEEE 118-node system. The system was divided into four regions, each containing conventional turbines and wind / solar power sources. Rolling scheduling was performed with a 15-minute step size and a 2-hour prediction window. The risk confidence level was [missing information]. Risk weights .

[0072] The implementation compared three schemes: distributed model predictive control, centralized model predictive control, and the two-level distributed model predictive control provided in this application. The main results include: tie-line operation: Under the distributed scheme, the tie-line frequently oscillated during the evening peak and repeatedly approached the 300 MW heat limit. This application obtained a smoother curve, basically maintaining within 90% of the heat limit, and the number of congestion violations decreased from 14 to 0, approaching the centralized benchmark; renewable energy utilization rate: The distributed utilization rate was about 82.3%, while this application improved it to 93.1% (close to the centralized 94.7%), corresponding to an additional absorption of 1200 MWh at the system level; computational efficiency and scalability: The average single-step solution time of this application was about 14.8 s, which increased linearly when the scale doubled, meeting real-time scheduling requirements; the centralized scheme averaged 38.6 s, with poor scalability; uncertainty risk: After introducing conditional risk values, the 95th percentile of the curtailment distribution converged significantly, and the heavy tail of the reserve shortage was compressed, that is, the probability density peak shifted to the low-risk side, verifying the risk suppression effect under rolling.

[0073] Based on this, the cross-regional new energy power system coordinated optimization method provided in this application, compared with existing centralized dispatch and decentralized autonomous methods, firstly, achieves hierarchical distributed coordinated optimization of the cross-regional power system by establishing a two-level structure of upper-level coordinator and lower-level regional controller. Unlike traditional centralized optimization, this architecture allows each region to perform optimization calculations independently, requiring only the exchange of a small amount of information such as boundary power and price to obtain a globally approximate optimal solution, thereby significantly reducing data communication volume and central computing pressure, and possessing good scalability; secondly, this application introduces a risk control mechanism based on conditional risk value in the lower-level model, which can probabilistically limit the impact of extreme new energy fluctuations on system security, realizing the quantification and dynamic adjustment of operational risks, compared with traditional robust optimization. Unlike conventional approaches, this application achieves a balance between economic efficiency and security through risk weighting coefficients, enabling the system to maintain high utilization of renewable energy while possessing stronger anti-disturbance capabilities. Furthermore, this application employs a rolling predictive control mechanism, using the latest load and renewable energy output forecasts for decision correction in each control cycle. This dynamically adapts to forecast errors and weather changes, improving the timeliness and flexibility of the dispatch strategy. Simultaneously, the alternating direction multiplier method between upper and lower layers achieves consistency and rapid convergence of boundary power, obtaining a stable solution within 10 to 15 iterations, fully meeting the real-time requirements of actual power dispatch. Finally, the framework of this application is compatible with multiple energy forms such as energy storage and integrated electric-thermal-gas energy systems, allowing for smooth integration into existing dispatch centers and enabling collaborative operation between different energy systems.

[0074] Simulation verification based on the IEEE 118-node system shows that this application can reduce the wind and solar curtailment rate by about 12% and increase the renewable energy utilization rate to over 93% without increasing system risk, with the average calculation time controlled within 15 seconds. Therefore, this application has significant engineering application value and promotion potential in terms of improving renewable energy absorption capacity, reducing backup configuration costs, and ensuring system operation safety.

[0075] Another aspect of the present invention provides a cross-regional new energy power system coordination and optimization device, such as... Figure 4As shown, the system includes a model building unit for constructing a two-level distributed optimization model, which comprises an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's reserve resource cost and cross-area tie-line transmission loss, while the optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk values. A collaborative iteration unit is used to perform collaborative iterative calculations between the upper-level coordinator and multiple area controllers using the alternating direction multiplier method within the rolling time domain of each control cycle. These collaborative iterative calculations include area update calculations, coordinator update calculations, and multiplier update calculations. An execution update unit is used to execute the intra-area scheduling scheme determined by the collaborative iterative calculations for the current moment when the collaborative iterative calculations meet preset convergence conditions, and to roll the time sequence forward by a preset step, performing collaborative iterative calculations for the next control cycle based on the updated system data.

[0076] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for coordinated optimization of cross-regional new energy power systems, characterized in that, include: A two-level distributed optimization model is constructed, wherein the two-level distributed optimization model includes an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's backup resource cost and cross-area tie line transmission loss. The optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk value. Within the rolling time domain of each control cycle, the upper-level coordinator performs collaborative iterative calculations between the multiple area controllers using the alternating direction multiplier method. The collaborative iterative calculations include area update calculations, coordinator update calculations, and multiplier update calculations. When the collaborative iterative calculation meets the preset convergence condition, the region executes the regional internal scheduling scheme determined by the collaborative iterative calculation for the current moment, and rolls the time sequence forward by a preset step, and performs collaborative iterative calculation for the next control cycle based on the updated system data.

2. The method according to claim 1, characterized in that, The constraints of the optimization model of the upper-level coordinator include at least the following: The sum of the power exchanged by tie lines between a region and all adjacent regions equals the net injected power of the region; the transmission power of tie lines is limited by capacity, and the power variation between adjacent time periods is limited by ramp rate; the power exchange between any two regions satisfies a reciprocal relationship, and the sum of the net injected power of the system is zero; the sum of the upper and lower reserve capacities provided by all regions is not less than the total reserve requirement of the system.

3. The method according to claim 1, characterized in that, The constraints of the optimization function of the area controller include at least the following: The sum of the power generation of all conventional generating units and available renewable energy sources in the region meets the needs of regional load and external interconnection power; the output of conventional generating units is within the preset range of technical output, and the output variation between adjacent time periods is limited by ramping capacity; the available output of renewable energy is decomposed into actual utilization power and planned curtailment power, and the planned curtailment power is a non-negative value.

4. The method according to claim 1, characterized in that, The conditional risk value is calculated in a linearized form, and the calculation process depends on a quantile variable and a set of non-negative relaxation variables corresponding to the new energy output scenario. The conditional risk value is represented as the sum of the quantile value and an adjustment term. The adjustment term is calculated based on the slack variables under different scenarios according to a preset confidence level. The slack variables under each scenario are constrained to be no less than the difference between the running loss under the corresponding scenario and the quantile value, while remaining non-negative.

5. The method according to claim 1, characterized in that, The regional update calculation includes: the regional controller, based on the current reference power and shadow price issued by the upper-layer coordinator, solves and updates the boundary power plan with the goal of minimizing the sum of the local optimization objective and the consistency penalty term; The coordinator update calculation includes: the upper-level coordinator aggregating the updated boundary power plans of all the area controllers, with the goal of minimizing the sum of the global optimization objective and the consistency penalty term, and solving and updating the reference power sent to the area; The multiplier update calculation includes: the upper-level coordinator updates the shadow price sent to the region based on a preset rule, according to the deviation between the updated boundary power plan of the region and its own updated reference power.

6. The method according to claim 1, characterized in that, The preset convergence conditions include: The original residual between the boundary power plan and the reference power of the region is lower than the first residual threshold, and the dual residual of the reference power change between adjacent iterations is lower than the second residual threshold.

7. The method according to claim 1, characterized in that, The region controller operates based on a rolling predictive control mechanism, wherein the rolling predictive control mechanism includes: At the beginning of each control cycle, the optimization function of the regional controller is calculated based on the load and renewable energy output forecast data within the forecast window starting from the current time. After obtaining the control instructions within the prediction window, the first control instruction corresponding to the current moment is implemented; The prediction window is rolled forward by a preset step, and the optimization and execution process is repeated in the next control cycle based on the updated system state and prediction data. At the same time, the updated price signal from the upper coordinator is received to form a continuous closed-loop optimization control.

8. The method according to claim 1, characterized in that, The penalty parameter in the alternating direction multiplier method is either a fixed value or dynamically adjusted based on the relative magnitude of the original residual and the dual residual.

9. The method according to any one of claims 1 to 8, characterized in that, When the region contains energy storage units, the constraints of the region controller include at least the following: The energy storage state of the energy storage unit is dynamically updated based on the charging power, discharging power and corresponding charging and discharging efficiency. The energy storage state of the energy storage unit is limited to between zero and its rated capacity; The charging power and discharging power of the energy storage unit are limited to between zero and the maximum power limit, respectively.

10. A cross-regional new energy power system coordination and optimization device, characterized in that, include: The model building unit is used to build a two-level distributed optimization model, wherein the two-level distributed optimization model includes an upper-level coordinator and multiple area controllers. The optimization objective of the upper-level coordinator is to minimize the system's backup resource cost and cross-area tie-line transmission loss. The optimization objective of the area controllers is to minimize the area's operating cost and risk loss based on conditional risk value. The collaborative iteration unit is used to perform collaborative iteration calculations between the upper-level coordinator and multiple region controllers in the rolling time domain of each control cycle by using the alternating direction multiplier method. The collaborative iteration calculations include region update calculations, coordinator update calculations, and multiplier update calculations. The execution update unit is used to execute the regional internal scheduling scheme determined by the collaborative iterative calculation for the current time when the collaborative iterative calculation meets the preset convergence condition, and to roll the time sequence forward by a preset step size, and to perform collaborative iterative calculation for the next control cycle based on the updated system data.

Images