Wind driven generator cluster power optimization distribution method and system

Abstract

The invention relates to the technical field of wind power generation cluster power optimization control. The invention provides a wind driven generator cluster power optimization distribution method and system. The method comprises the following steps: acquiring wind driven generator cluster basic data; based on the basic data, establishing a multi-target comprehensive optimization model; according to the multi-objective comprehensive optimization model, a weighted summation method is adopted to convert a multi-objective function into a single-objective function, and an optimization solution problem is constructed based on the power grid operation constraint conditions; performing iterative solution on the optimization solution problem by using a heuristic optimization algorithm to obtain an optimized power distribution scheme of each wind turbine generator in the cluster; and according to the optimized power distribution scheme, real-time power distribution control is carried out on a wind driven generator cluster through a cluster control system, and the fatigue load state of a unit is monitored to adjust a subsequent distribution strategy. The problem that an existing wind driven generator cluster power distribution method cannot give consideration to cluster economic benefits, unit fatigue load balance and operation stability at the same time is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power cluster power optimization control, in particular to a wind turbine cluster power optimization distribution method and system. BACKGROUND

[0002] With the increasing penetration of wind power in the power system, the demand for wind turbine clusters (including multiple wind farms or wind turbine groups) to participate in power regulation of the power grid is increasing. However, the volatility and uncertainty of wind power output pose challenges to cluster-level power optimization distribution. In the prior art, wind farm power distribution methods mainly focus on internal optimization of single stations or cluster-level economic dispatch. For example, patent document CN115833274A discloses a wind farm frequency modulation power distribution method considering historical fatigue load of units, which calculates the fatigue sensitivity coefficient of wind turbines, forms a scenario strategy set, and realizes the optimization distribution of active power within a single wind farm, aiming to reduce the accumulation of unit fatigue load and improve operational safety. On the other hand, patent document CN119231642A proposes a control method for a wind storage farm cluster, which uses a strategy game model (such as a zero determinant strategy) and a Shapley value benefit distribution mechanism to optimize the overall economic benefit of the cluster, and adjusts the unit control strategy through an inductive coefficient inversion model to realize the coordinated dispatch of the wind storage cluster.

[0003] Although the above-mentioned prior art has made progress at different levels, there are still obvious limitations. The method of CN115833274A focuses on the fatigue load management of units within a single wind farm and does not extend to coordinated distribution among multiple stations at the cluster level, and lacks consideration of the overall economic benefit of the cluster, making it difficult to adapt to the dispatching needs of large-scale wind power clusters. While CN119231642A involves cluster-level game optimization, its core focuses on economic benefit and voltage stability, without fully considering the historical fatigue state and mechanical load constraints of units, which may lead to accelerated wear of some units due to frequent adjustment, affecting the long-term operational reliability of the cluster. Therefore, the existing technology cannot simultaneously consider economic benefit and unit health status in cluster-level power distribution, and there is an urgent need for a power distribution method that can comprehensively optimize cluster benefit, fatigue load, and operational stability. SUMMARY

[0004] The present application aims to provide a wind turbine cluster power optimization distribution method and system, which aims to solve the problem that existing wind turbine cluster power distribution methods cannot simultaneously consider cluster economic benefit, unit fatigue load balance, and operational stability.

[0005] The present application is achieved by the following technical solutions:

[0006] A wind turbine cluster power optimization distribution method, comprising the following steps:

[0007] acquire wind turbine cluster basic data, the basic data including the unit parameters of each wind farm in the cluster, historical fatigue load data, cluster overall economic benefit parameters and power grid operation constraints;

[0008] Based on the basic data, a multi-objective comprehensive optimization model is established, which takes the maximization of cluster economic benefits, the equalization of unit fatigue loads and the optimization of operation stability as objective functions, and takes power demand constraints, unit mechanical load threshold constraints and cluster output response speed constraints as constraint conditions.

[0009] According to the multi-objective comprehensive optimization model, a weighted summation method is used to convert the multi-objective function into a single objective function, and an optimization solving problem is constructed based on the power grid operation constraints.

[0010] The heuristic optimization algorithm is used to iteratively solve the optimization solving problem to obtain an optimized power distribution scheme for each wind turbine unit in the cluster.

[0011] According to the optimized power distribution scheme, the wind turbine cluster is controlled by the cluster control system for real-time power distribution, and the unit fatigue load state is monitored to adjust the subsequent distribution strategy.

[0012] Optionally, the specific process of acquiring wind turbine cluster basic data is as follows:

[0013] Collect real-time operating parameters of each wind turbine unit in the cluster, including unit rated power, current output, wind speed prediction data and unit health status indicators.

[0014] Obtain historical fatigue load data of each wind turbine unit based on the unit monitoring system, and calculate the fatigue cumulative index and load distribution characteristics of each unit.

[0015] Obtain cluster overall economic benefit parameters from the power grid dispatching center or market trading platform, including time-of-use electricity price, power subsidy policy, operation cost coefficient and benefit optimization weight.

[0016] Obtain power grid operation constraints according to power grid dispatching instructions, including cluster total power demand, power regulation rate limit, frequency stability requirement and line transmission capacity constraint.

[0017] Integrate real-time operating parameters, historical fatigue load data, economic benefit parameters and power grid operation constraints to form a basic data set for subsequent establishment of a multi-objective comprehensive optimization model.

[0018] Optionally, the specific process of establishing a multi-objective comprehensive optimization model based on the basic data is as follows:

[0019] Based on the aforementioned basic dataset, the objective function input parameters and constraint input parameters are obtained by splitting the dataset. The objective function input parameters include the fatigue accumulation index of each unit, load distribution characteristics, time-of-use pricing, power subsidy policy, and unit response characteristic parameters. The constraint input parameters include the total power demand of the cluster, unit mechanical load threshold, power regulation rate limit, and line transmission capacity parameters.

[0020] Based on time-of-use pricing, power subsidy policies, and operating cost coefficients, an economic benefit objective function is constructed with the goal of maximizing the overall economic benefits of the cluster. Based on the historical fatigue accumulation index and load distribution characteristics of each unit, a fatigue load equalization function is constructed with the goal of minimizing the fatigue load variance of the units within the cluster. Based on grid frequency stability requirements, unit response speed, and line transmission capacity, an operational stability function is constructed with the goal of minimizing the output fluctuation deviation of the cluster.

[0021] Based on the power grid operation constraints and unit parameters, the constraint boundaries of the multi-objective integrated optimization model are set, including power grid power demand constraints, unit mechanical load threshold constraints, cluster output response speed limit constraints, and line transmission capacity over-limit constraints.

[0022] The economic benefit objective function, fatigue load equalization function, and operational stability function are used as parallel optimization objectives. The model is integrated with the aforementioned constraint boundaries to form a multi-objective comprehensive optimization model that simultaneously covers cluster benefits, unit health, and operational safety.

[0023] Optionally, the specific process of transforming the multi-objective function into a single-objective function using a weighted summation method based on the multi-objective comprehensive optimization model, and constructing the optimization problem based on the power grid operation constraints, is as follows:

[0024] Based on the objective function input parameters in the multi-objective integrated optimization model, the weight coefficients of each objective function are determined; wherein, the weight coefficients are dynamically configured according to the revenue optimization weight in the overall economic benefit parameters of the cluster, the fatigue accumulation index in the unit health status indicators, and the frequency stability requirements in the power grid operation constraints.

[0025] The economic benefit objective function, fatigue load equalization function, and operational stability function are linearly weighted and combined using a weighted summation method to form a comprehensive objective function; wherein, the comprehensive objective function is expressed as a weighted sum of the objective functions, and the sum of the weight coefficients is 1;

[0026] Based on the total power demand of the power grid cluster, the mechanical load threshold of the generating units, the power regulation rate limit, and the line transmission capacity constraint in the power grid operation constraints, a set of constraints for the optimization problem is constructed, including equality constraints and inequality constraints.

[0027] The comprehensive objective function is integrated with the set of constraints to define a single-objective optimization problem with the power allocation of each wind turbine in the cluster as the decision variable, which is then used to solve the problem using a subsequent heuristic optimization algorithm.

[0028] Optionally, the specific process of using a heuristic optimization algorithm to iteratively solve the optimization problem and obtain the optimized power allocation scheme for each wind turbine in the cluster is as follows:

[0029] Based on the single-objective optimization problem, initialize the parameters of the heuristic optimization algorithm, including population size, maximum number of iterations, and mutation probability;

[0030] An initial population is randomly generated based on the power allocation range of each wind turbine in the cluster; where each individual represents a preset power allocation scheme.

[0031] Based on the comprehensive objective function and the set of constraints, the fitness value of each individual is calculated, and individuals that violate the constraints are handled by the penalty function method.

[0032] Based on the fitness value, selection operations are used to select superior individuals from the current population, and crossover and mutation operations are used to generate a new population to explore the solution space.

[0033] The population is iteratively updated until the maximum number of iterations or the fitness value convergence threshold is reached, and the best individual in each generation is recorded.

[0034] The optimal power allocation value for the best individual is extracted from the final population and used as the optimized power allocation scheme for each wind turbine in the cluster.

[0035] Optionally, the specific process of performing real-time power allocation control of the wind turbine cluster through the cluster control system according to the optimized power allocation scheme, and monitoring the fatigue load status of the units to adjust the subsequent allocation strategy, is as follows:

[0036] Based on the optimized power allocation scheme, power control commands for each wind turbine in the cluster are generated and sent to the execution units of each turbine through the cluster control system for real-time power allocation control.

[0037] Real-time collection of operating status data for each wind turbine, including current output, mechanical load parameters and fatigue accumulation index, and updating of historical fatigue load data based on the turbine monitoring system;

[0038] The fatigue cumulative index collected in real time is compared with historical fatigue load data to calculate the fatigue load change rate, and the risk of unit health status degradation is assessed in combination with unit health status indicators.

[0039] Based on the evaluation results, the weight coefficients or constraint parameters of the objective function in the multi-objective integrated optimization model are dynamically adjusted; wherein, the weight coefficients are reconfigured according to the fatigue load change rate and the frequency stability requirements in the power grid operation constraints.

[0040] Based on the adjusted multi-objective integrated optimization model, the optimization solution process is re-executed to generate an updated optimized power allocation scheme, and cyclic power allocation control is implemented through the cluster control system.

[0041] Optionally, the specific process of comparing the real-time collected fatigue accumulation index with historical fatigue load data, calculating the fatigue load change rate, and combining it with unit health status indicators to assess the risk of unit health status degradation is as follows:

[0042] The system reads real-time fatigue accumulation index and stored historical fatigue load data from the unit monitoring system; the historical fatigue load data includes the fatigue accumulation index time series within a preset time period in the past.

[0043] Based on the fatigue cumulative index time series, the deviation rate between the current fatigue cumulative index and the historical average is calculated using the sliding window method, or the slope of the fatigue cumulative index over time is calculated using the linear regression method to obtain the fatigue load change rate.

[0044] Obtain unit health status indicators from the unit health status database, including the unit's cumulative operating hours, the time of the most recent maintenance, the number of historical failures, and the wear parameters of key mechanical components;

[0045] The fatigue load change rate and the unit health status index are input into a preset risk assessment model to calculate the unit health status degradation risk value; wherein, the risk assessment model is a linear weighted model or a machine learning model, and the unit health status degradation risk value output by the risk assessment model is positively correlated with the fatigue load change rate and negatively correlated with the unit health status index;

[0046] The risk level of unit health status degradation is determined based on whether the risk value exceeds the preset risk threshold. This information is then used to optimize model parameter adjustments and generate risk warning signals.

[0047] Based on the same inventive concept, the present invention also provides a wind turbine cluster power optimization allocation system for implementing the aforementioned wind turbine cluster power optimization allocation method, comprising:

[0048] The data acquisition module is used to acquire basic data of the wind turbine cluster, including unit parameters of each wind farm in the cluster, historical fatigue load data, overall economic benefit parameters of the cluster, and grid operation constraints.

[0049] The multi-objective integrated optimization model establishment module is used to establish a multi-objective integrated optimization model based on the aforementioned basic data. The objective functions are maximizing the economic benefits of the cluster, equalizing the fatigue load of the units, and optimizing the operational stability. The constraints are grid power demand constraints, unit mechanical load threshold constraints, and cluster output response speed constraints.

[0050] The optimization problem construction module is used to transform the multi-objective function into a single-objective function using the weighted summation method based on the multi-objective comprehensive optimization model, and to construct the optimization problem based on the power grid operation constraints.

[0051] The heuristic optimization solution module is used to iteratively solve the optimization problem using a heuristic optimization algorithm to obtain the optimal power allocation scheme for each wind turbine unit in the cluster.

[0052] The power distribution control module is used to perform real-time power distribution control on the wind turbine cluster through the cluster control system according to the optimized power distribution scheme, and to monitor the fatigue load status of the units to adjust the subsequent distribution strategy.

[0053] Based on the same inventive concept, the present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-described wind turbine cluster power optimization allocation method.

[0054] Based on the same inventive concept, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for optimizing power allocation of wind turbine clusters.

[0055] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0056] By establishing a multi-objective comprehensive optimization model that simultaneously covers maximizing cluster economic benefits, equalizing unit fatigue loads, and optimizing operational stability, the limitations of single optimization objectives in existing technologies are overcome. This model can effectively reduce the accumulation of fatigue loads on key units, extend equipment service life, and achieve dual protection of economic benefits and unit health status while meeting the needs of power grid dispatch.

[0057] By introducing historical fatigue load data and mechanical load threshold constraints of the units, the actual load-bearing capacity and historical loss of the equipment are fully considered during the power allocation process, avoiding the aggravation of equipment damage caused by excessive adjustment or frequent start-stop. Combined with the optimization of the operation stability target, the operational reliability and disturbance resistance of the wind power cluster are significantly enhanced throughout its entire life cycle.

[0058] The model is solved by using a heuristic optimization algorithm, which can efficiently handle complex optimization problems with multiple constraints and nonlinearity, adapt to the fluctuations and uncertainties of wind power output, and form a closed-loop optimization mechanism by monitoring the fatigue status of the units in real time and dynamically adjusting the allocation strategy. This enables the system to have good response speed and adaptive capability, and meet the grid's requirements for rapid dispatch of wind power clusters. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating the wind turbine cluster power optimization allocation method according to an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the wind turbine cluster power optimization allocation system according to an embodiment of the present invention. Detailed Implementation

[0061] The following is a detailed description of the embodiments, in conjunction with the accompanying drawings.

[0062] Reference Figure 1 A method for optimizing power allocation in a wind turbine cluster includes the following steps:

[0063] Step 1: Obtain basic data of the wind turbine cluster. The basic data includes the unit parameters of each wind farm in the cluster, historical fatigue load data, overall economic benefit parameters of the cluster, and grid operation constraints.

[0064] In some embodiments, the specific process of obtaining basic data of the wind turbine cluster is as follows:

[0065] Collect real-time operating parameters of each wind turbine in the cluster, including rated power, current output, wind speed forecast data, and unit health status indicators;

[0066] Historical fatigue load data of each wind turbine is obtained based on the unit monitoring system, and the fatigue accumulation index and load distribution characteristics of each unit are calculated.

[0067] Obtain the overall economic benefit parameters of the cluster from the power grid dispatch center or market trading platform, including time-of-use pricing, power subsidy policies, operating cost coefficients, and benefit optimization weights.

[0068] Obtain grid operation constraints based on grid dispatch instructions, including total power demand of the cluster, power regulation rate limit, frequency stability requirements and line transmission capacity constraints;

[0069] By integrating real-time operating parameters, historical fatigue load data, economic benefit parameters, and power grid operation constraints, a basic dataset is formed for the subsequent establishment of a multi-objective integrated optimization model.

[0070] Step 2: Based on the aforementioned basic data, establish a multi-objective comprehensive optimization model. The multi-objective comprehensive optimization model takes maximizing the economic benefits of the cluster, equalizing the fatigue load of the units, and optimizing the operational stability as the objective functions, and takes the power demand constraints of the power grid, the mechanical load threshold constraints of the units, and the output response speed constraints of the cluster as the constraints.

[0071] In some embodiments, the specific process of establishing a multi-objective comprehensive optimization model based on the basic data is as follows:

[0072] Based on the aforementioned basic dataset, the objective function input parameters and constraint input parameters are obtained. The objective function input parameters include the fatigue accumulation index of each unit, load distribution characteristics, time-of-use pricing, power subsidy policies, and unit response characteristic parameters. The constraint input parameters include the total power demand of the cluster, unit mechanical load thresholds, power regulation rate limits, and line transmission capacity parameters. Assume the wind turbine cluster has... A wind turbine unit, indexed as The decision variable is the allocated power of each unit.

[0073] Based on time-of-use pricing, power subsidy policies, and operating cost coefficients, an economic benefit objective function is constructed with the goal of maximizing the overall economic benefit of the cluster. This objective function can be represented by the following equation (1):

[0074]

[0075] in, The overall economic benefit of the cluster represents the total economic benefit obtained by the cluster after power allocation. For the first The revenue coefficient of an individual generating unit, including factors such as time-of-use pricing and power subsidy policies, represents the economic revenue generated per unit of power. For the first The allocated power (decision variable) of each unit represents the active power that the unit needs to output after optimization.

[0076] In the optimization process, it is transformed into a minimization problem, so the minimization objective function is defined as shown in equation (2):

[0077]

[0078] in, Let be the transformed economic benefit objective function. Since optimization algorithms typically deal with minimization problems, maximizing economic benefits is transformed into minimizing negative benefits.

[0079] Based on the historical fatigue accumulation index and load distribution characteristics of each unit, a fatigue load equalization function is constructed with the objective of minimizing the fatigue load variance of units within the cluster. The function can be defined with the objective of minimizing the fatigue load variance of units within the cluster, defining the ... The fatigue load of each unit is The variance of the fatigue load is shown in equation (3) below:

[0080]

[0081] in, This represents the variance of fatigue loads across all units within the cluster, used to measure the degree of balance in fatigue loads (a smaller value indicates a more balanced load). For the first The fatigue load of a unit after power allocation is a combined value of historical fatigue accumulation and the impact of current power allocation; The average fatigue load of all units within the cluster; For the first The historical fatigue accumulation index of each unit is calculated based on historical fatigue load data and reflects the accumulated fatigue damage of the unit. For the first The load distribution characteristic parameters of each unit represent the influence coefficient of unit power change on fatigue load.

[0082] Because the power demand constraint of the power grid fixes the total power, therefore Can be written as a constant term and The function is shown in equation (4) below:

[0083]

[0084] in, This is the average of the historical cumulative fatigue index of all units within the cluster; This represents the average value of the load distribution characteristic parameters of all units within the cluster; The total power demand of the cluster comes from the grid dispatch command and represents the total active power that the cluster needs to output.

[0085] The fatigue load equalization function minimizes the variance, as shown in equation (5) below:

[0086]

[0087] in, The objective function for fatigue load equalization is to minimize the fatigue load variance in order to achieve a balanced distribution of fatigue loads among units.

[0088] Based on the requirements of grid frequency stability, unit response speed, and line transmission capacity, an operational stability function is constructed with the goal of minimizing the power output fluctuation deviation of the power cluster. This function can be defined as minimizing the sum of squares of the deviations between the allocated power and the current output of each unit, as shown in equation (6).

[0089]

[0090] in, To improve operational stability, the objective function is to minimize the cluster output fluctuation deviation. For the first The current output of a unit represents the actual output power of that unit before optimization.

[0091] Based on the power grid operation constraints and unit parameters, the constraint boundaries of the multi-objective integrated optimization model are set, including power grid demand constraints, unit mechanical load threshold constraints, cluster output response speed limits, and line transmission capacity over-limit constraints. The power grid demand constraints are shown in equation (7) below:

[0092]

[0093] The mechanical load threshold constraint condition of the unit is shown in the following equation (8):

[0094]

[0095] in, For the first The minimum allowable power of each unit is determined by the unit's mechanical load threshold. For the first The maximum allowable power of each unit is determined by the unit's mechanical load threshold. This indicates that the constraint applies to each unit. Both are true;

[0096] The constraint condition for the output response speed of the cluster is shown in equation (9) below:

[0097]

[0098] in, For the first The maximum allowable power regulation of a unit represents the maximum limit of power change of the unit per unit time, used to ensure that the response speed does not exceed the unit's capacity.

[0099] The line transmission capacity over-limit constraint is shown in equation (10):

[0100]

[0101] in, This is an index for lines in the power grid; the range of values ​​is determined based on the actual number of lines. For the first The unit is for the first The power transmission impact coefficient of the line represents the unit's Power changes on the line The extent of the influence of trends; For the first The transmission capacity limit of each line; This indicates that the constraint applies to each line. Both are valid.

[0102] The economic benefit objective function, fatigue load equalization function, and operational stability function are used as parallel optimization objectives. The model is integrated with the aforementioned constraint boundaries to form a multi-objective comprehensive optimization model that simultaneously covers cluster benefits, unit health, and operational safety.

[0103] Step 3: Based on the multi-objective integrated optimization model, the multi-objective function is transformed into a single-objective function using the weighted summation method, and an optimization problem is constructed based on the power grid operation constraints.

[0104] In some embodiments, the specific process of transforming the multi-objective function into a single-objective function using a weighted summation method based on the multi-objective comprehensive optimization model, and constructing the optimization problem based on the power grid operation constraints, is as follows:

[0105] Based on the objective function input parameters in the multi-objective integrated optimization model, the weight coefficients of each objective function are determined; wherein, the weight coefficients are dynamically configured according to the revenue optimization weight in the overall economic benefit parameters of the cluster, the fatigue accumulation index in the unit health status indicators, and the frequency stability requirements in the power grid operation constraints.

[0106] The economic benefit objective function, fatigue load equalization function, and operational stability function are linearly weighted and combined using a weighted summation method to form a comprehensive objective function; wherein, the comprehensive objective function is expressed as a weighted sum of the objective functions, and the sum of the weight coefficients is 1;

[0107] Based on the total power demand of the power grid cluster, the mechanical load threshold of the generating units, the power regulation rate limit, and the line transmission capacity constraint in the power grid operation constraints, a set of constraints for the optimization problem is constructed, including equality constraints and inequality constraints.

[0108] The comprehensive objective function is integrated with the set of constraints to define a single-objective optimization problem with the power allocation of each wind turbine in the cluster as the decision variable, which is then used to solve the problem using a subsequent heuristic optimization algorithm.

[0109] Step 4: Use a heuristic optimization algorithm to iteratively solve the optimization problem and obtain the optimal power allocation scheme for each wind turbine in the cluster.

[0110] In some embodiments, the specific process of iteratively solving the optimization problem using a heuristic optimization algorithm to obtain the optimized power allocation scheme for each wind turbine in the cluster is as follows:

[0111] Based on the single-objective optimization problem, initialize the parameters of the heuristic optimization algorithm, including population size, maximum number of iterations, and mutation probability;

[0112] An initial population is randomly generated based on the power allocation range of each wind turbine in the cluster; where each individual represents a preset power allocation scheme.

[0113] Based on the comprehensive objective function and the set of constraints, the fitness value of each individual is calculated, and individuals that violate the constraints are handled by the penalty function method.

[0114] Based on the fitness value, selection operations are used to select superior individuals from the current population, and crossover and mutation operations are used to generate a new population to explore the solution space.

[0115] The population is iteratively updated until the maximum number of iterations or the fitness value convergence threshold is reached, and the best individual in each generation is recorded.

[0116] The optimal power allocation value for the best individual is extracted from the final population and used as the optimized power allocation scheme for each wind turbine in the cluster.

[0117] Step 5: Based on the optimized power allocation scheme, the wind turbine cluster is controlled in real time through the cluster control system, and the fatigue load status of the units is monitored to adjust the subsequent allocation strategy.

[0118] In some embodiments, the specific process of performing real-time power allocation control on the wind turbine cluster through the cluster control system according to the optimized power allocation scheme, and monitoring the fatigue load status of the units to adjust the subsequent allocation strategy is as follows:

[0119] Based on the optimized power allocation scheme, power control commands for each wind turbine in the cluster are generated and sent to the execution units of each turbine through the cluster control system for real-time power allocation control.

[0120] Real-time collection of operating status data for each wind turbine, including current output, mechanical load parameters and fatigue accumulation index, and updating of historical fatigue load data based on the turbine monitoring system;

[0121] The fatigue cumulative index collected in real time is compared with historical fatigue load data to calculate the fatigue load change rate, and the risk of unit health status degradation is assessed in combination with unit health status indicators.

[0122] Based on the evaluation results, the weight coefficients or constraint parameters of the objective function in the multi-objective integrated optimization model are dynamically adjusted; wherein, the weight coefficients are reconfigured according to the fatigue load change rate and the frequency stability requirements in the power grid operation constraints.

[0123] Based on the adjusted multi-objective integrated optimization model, the optimization solution process is re-executed to generate an updated optimized power allocation scheme, and cyclic power allocation control is implemented through the cluster control system.

[0124] In some embodiments, the specific process of comparing the real-time collected fatigue accumulation index with historical fatigue load data, calculating the fatigue load change rate, and combining it with unit health status indicators to assess the risk of unit health status degradation is as follows:

[0125] The system reads real-time fatigue accumulation index and stored historical fatigue load data from the unit monitoring system; the historical fatigue load data includes the fatigue accumulation index time series within a preset time period in the past.

[0126] Based on the fatigue cumulative index time series, the deviation rate between the current fatigue cumulative index and the historical average is calculated using the sliding window method, or the slope of the fatigue cumulative index over time is calculated using the linear regression method to obtain the fatigue load change rate.

[0127] Obtain unit health status indicators from the unit health status database, including the unit's cumulative operating hours, the time of the most recent maintenance, the number of historical failures, and the wear parameters of key mechanical components;

[0128] The fatigue load change rate and unit health status indicators are input into a preset risk assessment model to calculate the unit health status degradation risk value. The risk assessment model can be a linear weighted model or a machine learning model. The unit health status degradation risk value output by the risk assessment model is positively correlated with the fatigue load change rate and negatively correlated with the unit health status indicators. For example, if the risk assessment model is a linear weighted model, the unit health status degradation risk value can be set as follows: The rate of change of fatigue load The unit's health status indicators are .in, It is a comprehensive value obtained by weighted combination of multiple health status parameters, and A larger value indicates a better health status of the unit. The linear weighted model is then shown in equation (11):

[0129]

[0130] in, and These are positive weighting coefficients used to adjust the contribution of fatigue load change rate and comprehensive health status index to the risk value. The weighting coefficients can be dynamically configured according to actual operational needs to meet [the requirements]. Or other normalization conditions;

[0131] The fatigue load change rate can be calculated using the following two methods:

[0132] Deviation rate method: The deviation rate between the current fatigue cumulative index and the historical average is calculated based on the sliding window, as shown in the following formula (12):

[0133]

[0134] in, This represents the current cumulative fatigue index. This is the historical average of the cumulative fatigue index over a preset time period.

[0135] The slope of change method: The slope of the fatigue accumulation index over time is calculated using the linear regression method, as shown in the following formula (13):

[0136]

[0137] in, For time points The fatigue accumulation index; It is the average value at different points in time. This is the size of the sliding window.

[0138] The health status index of the unit is calculated as shown in the following formula (14):

[0139]

[0140] in:

[0141] A health score based on cumulative operating hours can be defined as follows: , This represents the cumulative operating hours. The design lifespan in hours (a higher normalized value indicates better health);

[0142] The health score based on the most recent maintenance time can be defined as follows: , This is the number of hours since the last maintenance. Standard maintenance interval hours;

[0143] The health score for the number of historical failures can be defined as follows: ,in This represents the number of historical failures. The maximum number of failures allowed;

[0144] The health score for the wear of key mechanical components can be defined as follows: ,in For wear parameters, Maximum permissible wear;

[0145] The weighting coefficients for each health score satisfy the following conditions: The weights can be configured based on the unit type and operating experience.

[0146] The risk level of unit health status degradation is determined based on whether the risk value exceeds the preset risk threshold. This information is then used to optimize model parameter adjustments and generate risk warning signals.

[0147] Based on the same inventive concept, and corresponding to any of the above embodiments, refer to... Figure 2 This invention provides a wind turbine cluster power optimization allocation system for implementing the aforementioned wind turbine cluster power optimization allocation method, comprising:

[0148] The data acquisition module is used to acquire basic data of the wind turbine cluster, including unit parameters of each wind farm in the cluster, historical fatigue load data, overall economic benefit parameters of the cluster, and grid operation constraints.

[0149] The multi-objective integrated optimization model establishment module is used to establish a multi-objective integrated optimization model based on the aforementioned basic data. The objective functions are maximizing the economic benefits of the cluster, equalizing the fatigue load of the units, and optimizing the operational stability. The constraints are grid power demand constraints, unit mechanical load threshold constraints, and cluster output response speed constraints.

[0150] The optimization problem construction module is used to transform the multi-objective function into a single-objective function using the weighted summation method based on the multi-objective comprehensive optimization model, and to construct the optimization problem based on the power grid operation constraints.

[0151] The heuristic optimization solution module is used to iteratively solve the optimization problem using a heuristic optimization algorithm to obtain the optimal power allocation scheme for each wind turbine unit in the cluster.

[0152] The power distribution control module is used to perform real-time power distribution control on the wind turbine cluster through the cluster control system according to the optimized power distribution scheme, and to monitor the fatigue load status of the units to adjust the subsequent distribution strategy.

[0153] Based on the same inventive concept, corresponding to any of the above embodiments, the present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the wind turbine cluster power optimization allocation method of the embodiment.

[0154] Alternatively, the aforementioned electronic device may be a server.

[0155] In addition, this embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the wind turbine cluster power optimization allocation method of the embodiment.

[0156] It is understood that the processor in the embodiments of the present invention may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0157] The method steps in the embodiments of the present invention can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0158] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a storage medium or transmitted through a storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive (SSD)).

Claims

1. A method for optimizing power allocation in a wind turbine cluster, characterized in that, Includes the following steps: Acquire basic data of the wind turbine cluster, including unit parameters of each wind farm in the cluster, historical fatigue load data, overall economic benefit parameters of the cluster, and grid operation constraints; Based on the aforementioned basic data, a multi-objective comprehensive optimization model is established. The multi-objective comprehensive optimization model takes maximizing the economic benefits of the cluster, equalizing the fatigue load of the units, and optimizing the operational stability as the objective functions, and takes grid power demand constraints, unit mechanical load threshold constraints, and cluster output response speed constraints as the constraints. Based on the multi-objective integrated optimization model, the multi-objective function is transformed into a single-objective function using the weighted summation method, and an optimization problem is constructed based on the power grid operation constraints. The optimization problem is solved iteratively using a heuristic optimization algorithm to obtain an optimized power allocation scheme for each wind turbine unit in the cluster. According to the optimized power allocation scheme, the wind turbine cluster is controlled in real time by the cluster control system, and the fatigue load status of the units is monitored to adjust the subsequent allocation strategy.

2. The wind turbine cluster power optimization allocation method as described in claim 1, characterized in that... The specific process for obtaining the basic data of the wind turbine cluster is as follows: Collect real-time operating parameters of each wind turbine in the cluster, including rated power, current output, wind speed forecast data, and unit health status indicators; Historical fatigue load data of each wind turbine is obtained based on the unit monitoring system, and the fatigue accumulation index and load distribution characteristics of each unit are calculated. Obtain the overall economic benefit parameters of the cluster from the power grid dispatch center or market trading platform, including time-of-use pricing, power subsidy policies, operating cost coefficients, and benefit optimization weights. Obtain grid operation constraints based on grid dispatch instructions, including total power demand of the cluster, power regulation rate limit, frequency stability requirements and line transmission capacity constraints; By integrating real-time operating parameters, historical fatigue load data, economic benefit parameters, and power grid operation constraints, a basic dataset is formed for the subsequent establishment of a multi-objective integrated optimization model.

3. The wind turbine cluster power optimization allocation method as described in claim 1, characterized in that, The specific process of establishing a multi-objective comprehensive optimization model based on the aforementioned basic data is as follows: Based on the aforementioned basic dataset, the objective function input parameters and constraint input parameters are obtained by splitting the dataset. The objective function input parameters include the fatigue accumulation index of each unit, load distribution characteristics, time-of-use pricing, power subsidy policy, and unit response characteristic parameters. The constraint input parameters include the total power demand of the cluster, unit mechanical load threshold, power regulation rate limit, and line transmission capacity parameters. Based on time-of-use pricing, power subsidy policies, and operating cost coefficients, an economic benefit objective function is constructed with the goal of maximizing the overall economic benefits of the cluster. Based on the historical fatigue accumulation index and load distribution characteristics of each unit, a fatigue load equalization function is constructed with the goal of minimizing the fatigue load variance of the units within the cluster. Based on grid frequency stability requirements, unit response speed, and line transmission capacity, an operational stability function is constructed with the goal of minimizing the output fluctuation deviation of the cluster. Based on the power grid operation constraints and unit parameters, the constraint boundaries of the multi-objective integrated optimization model are set, including power grid power demand constraints, unit mechanical load threshold constraints, cluster output response speed limit constraints, and line transmission capacity over-limit constraints. The economic benefit objective function, fatigue load equalization function, and operational stability function are used as parallel optimization objectives. The model is integrated with the aforementioned constraint boundaries to form a multi-objective comprehensive optimization model that simultaneously covers cluster benefits, unit health, and operational safety.

4. The wind turbine cluster power optimization allocation method as described in claim 1, characterized in that, The specific process of transforming the multi-objective function into a single-objective function using the weighted summation method based on the multi-objective comprehensive optimization model, and constructing the optimization problem based on the power grid operation constraints, is as follows: Based on the objective function input parameters in the multi-objective integrated optimization model, the weight coefficients of each objective function are determined; wherein, the weight coefficients are dynamically configured according to the revenue optimization weight in the overall economic benefit parameters of the cluster, the fatigue accumulation index in the unit health status indicators, and the frequency stability requirements in the power grid operation constraints. The economic benefit objective function, fatigue load equalization function, and operational stability function are linearly weighted and combined using a weighted summation method to form a comprehensive objective function; wherein, the comprehensive objective function is expressed as a weighted sum of the objective functions, and the sum of the weight coefficients is 1; Based on the total power demand of the cluster, the mechanical load threshold of the generating units, the power regulation rate limit, and the line transmission capacity constraint in the power grid operation constraints, a set of constraints for the optimization problem is constructed, including equality constraints and inequality constraints. The comprehensive objective function is integrated with the set of constraints to define a single-objective optimization problem with the power allocation of each wind turbine in the cluster as the decision variable, which is then used to solve the problem using a subsequent heuristic optimization algorithm.

5. The wind turbine cluster power optimization allocation method as described in claim 4, characterized in that, The specific process of using a heuristic optimization algorithm to iteratively solve the optimization problem and obtain the optimal power allocation scheme for each wind turbine in the cluster is as follows: Based on the single-objective optimization problem, initialize the parameters of the heuristic optimization algorithm, including population size, maximum number of iterations, and mutation probability; An initial population is randomly generated based on the power allocation range of each wind turbine in the cluster; where each individual represents a preset power allocation scheme. Based on the comprehensive objective function and the set of constraints, the fitness value of each individual is calculated, and individuals that violate the constraints are handled by the penalty function method. Based on the fitness value, selection operations are used to select superior individuals from the current population, and crossover and mutation operations are used to generate a new population to explore the solution space. The population is iteratively updated until the maximum number of iterations or the fitness value convergence threshold is reached, and the best individual in each generation is recorded. The optimal power allocation value for the best individual is extracted from the final population and used as the optimized power allocation scheme for each wind turbine in the cluster.

6. The wind turbine cluster power optimization allocation method as described in claim 1, characterized in that, The specific process of performing real-time power allocation control of the wind turbine cluster through the cluster control system according to the optimized power allocation scheme, and monitoring the fatigue load status of the units to adjust the subsequent allocation strategy is as follows: Based on the optimized power allocation scheme, power control commands for each wind turbine in the cluster are generated and sent to the execution units of each turbine through the cluster control system for real-time power allocation control. Real-time collection of operating status data for each wind turbine, including current output, mechanical load parameters and fatigue accumulation index, and updating of historical fatigue load data based on the turbine monitoring system; The fatigue cumulative index collected in real time is compared with historical fatigue load data to calculate the fatigue load change rate, and the risk of unit health status degradation is assessed in combination with unit health status indicators. Based on the evaluation results, the weight coefficients or constraint parameters of the objective function in the multi-objective integrated optimization model are dynamically adjusted; wherein, the weight coefficients are reconfigured according to the fatigue load change rate and the frequency stability requirements in the power grid operation constraints. Based on the adjusted multi-objective integrated optimization model, the optimization solution process is re-executed to generate an updated optimized power allocation scheme, and cyclic power allocation control is implemented through the cluster control system.

7. The wind turbine cluster power optimization allocation method as described in claim 6, characterized in that, The specific process of comparing the real-time collected fatigue accumulation index with historical fatigue load data, calculating the fatigue load change rate, and combining it with unit health status indicators to assess the risk of unit health status degradation is as follows: The system reads real-time fatigue accumulation index and stored historical fatigue load data from the unit monitoring system; the historical fatigue load data includes the fatigue accumulation index time series within a preset time period in the past. Based on the fatigue cumulative index time series, the deviation rate between the current fatigue cumulative index and the historical average is calculated using the sliding window method, or the slope of the fatigue cumulative index over time is calculated using the linear regression method to obtain the fatigue load change rate. Obtain unit health status indicators from the unit health status database, including the unit's cumulative operating hours, the time of the most recent maintenance, the number of historical failures, and the wear parameters of key mechanical components; The fatigue load change rate and the unit health status index are input into a preset risk assessment model to calculate the unit health status degradation risk value; wherein, the risk assessment model is a linear weighted model or a machine learning model, and the unit health status degradation risk value output by the risk assessment model is positively correlated with the fatigue load change rate and negatively correlated with the unit health status index; The risk level of unit health status degradation is determined based on whether the risk value exceeds the preset risk threshold. This information is then used to optimize model parameter adjustments and generate risk warning signals.

8. A wind turbine cluster power optimization allocation system, used to implement the wind turbine cluster power optimization allocation method according to any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire basic data of the wind turbine cluster, including unit parameters of each wind farm in the cluster, historical fatigue load data, overall economic benefit parameters of the cluster, and grid operation constraints. The multi-objective integrated optimization model establishment module is used to establish a multi-objective integrated optimization model based on the aforementioned basic data. The objective functions are maximizing the economic benefits of the cluster, equalizing the fatigue load of the units, and optimizing the operational stability. The constraints are grid power demand constraints, unit mechanical load threshold constraints, and cluster output response speed constraints. The optimization problem construction module is used to transform the multi-objective function into a single-objective function using the weighted summation method based on the multi-objective comprehensive optimization model, and to construct the optimization problem based on the power grid operation constraints. The heuristic optimization solution module is used to iteratively solve the optimization problem using a heuristic optimization algorithm to obtain the optimal power allocation scheme for each wind turbine unit in the cluster. The power distribution control module is used to perform real-time power distribution control on the wind turbine cluster through the cluster control system according to the optimized power distribution scheme, and to monitor the fatigue load status of the units to adjust the subsequent distribution strategy.

9. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the wind turbine cluster power optimization allocation method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the wind turbine cluster power optimization allocation method as described in any one of claims 1-7.

Images