Intelligent power supply scheduling method and system of self-adaptive load

Abstract

The method comprises the following steps: performing cluster division based on cut-in and cut-out wind speed and regional wind speed distribution of a wind turbine generator, performing operation characteristic decoupling of different time frequency domains on each cluster, constructing a frequency modulation inertia model, and on the premise of meeting a power grid load demand, performing frequency modulation on the frequency modulation inertia model to obtain a frequency modulation inertia model; performing multi-objective optimization by taking the current state of the cluster and the frequency modulation model as constraints, allocating the optimal frequency modulation task to each cluster, then considering the synergistic effect between the clusters, optimizing the scheduling task in real time, and determining a start-stop time scheme set of the wind turbine generator in each cluster by taking power balance as a condition; and the scheme with the minimum power peak-valley difference is selected as a final start-stop method. According to the method, high-efficiency operation of the wind power plant and stable power supply of the power grid are realized through cluster division and multi-objective optimization scheduling based on the characteristics of the wind turbine generator and regional wind speed distribution, the wind curtailment phenomenon and the operation cost of the wind power plant are reduced, the economic benefit is improved, and the method has relatively good interpretability.

Description

Technical Field

[0001] The present invention relates to the field of electric power dispatching, and in particular to an adaptive load intelligent power dispatching method and system. Background Art

[0002] In recent years, wind power generation has played an increasingly important role in the global energy structure, and its installed capacity has continued to expand. However, wind power generation is extremely intermittent and volatile, and wind speed is affected by complex meteorological factors and changes randomly and dramatically, making it difficult to accurately estimate the output power of wind turbines.

[0003] Frequent fluctuations in wind turbine power can cause instability in grid frequency and voltage. At the same time, the output power and operating frequency of wind turbines are susceptible to external interference and fluctuations. Traditional dispatching lacks in-depth analysis of its operating characteristics in different time and frequency domains, and cannot effectively decouple these factors, making it difficult to build an accurate inertial model of the frequency regulation capability of wind turbine clusters. As a result, when the grid load fluctuates, it is difficult to accurately regulate the frequency regulation task of wind turbine clusters, and it is impossible to meet the grid's strict requirements for power stability. Moreover, the distribution of wind resources in different regions varies greatly, and the construction of wind farms is restricted by geographical conditions. During dispatching, it is necessary to comprehensively coordinate the cooperation between multiple wind farms, and traditional dispatching methods are difficult to achieve effective coordination.

[0004] An adaptive load intelligent power dispatching method and system ensures the safe and stable operation of the power grid. It accurately estimates the power of wind turbines based on complex meteorological data, deeply analyzes the time-frequency domain characteristics, builds an accurate inertia model, achieves efficient coordination of multiple wind farms, effectively stabilizes the frequency and voltage of the power grid, ensures a stable power supply, and ensures efficient and stable operation of the power system. Summary of the invention

[0005] The object of the present invention is to provide an intelligent power supply scheduling method and system for adaptive loads.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] A first aspect of the present invention provides an intelligent power scheduling method for adaptive loads, comprising:

[0008] Wind turbines are clustered based on their cut-in and cut-out wind speeds and regional wind speed distribution;

[0009] Decouple the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and build an inertial model of the frequency regulation capability of the wind turbine cluster;

[0010] Under the condition of meeting the predicted load demand of the power grid, multi-objective optimization is performed with the current operating status of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, and the frequency regulation tasks corresponding to the optimal solution are allocated to each wind turbine cluster;

[0011] Considering the synergy between wind turbine clusters, the scheduling task is optimized in real time;

[0012] In each wind turbine cluster, a start and stop time plan set for each wind turbine in different time periods is determined according to the scheduling task, subject to satisfying the power balance of each wind turbine in the cluster;

[0013] A scheme with the smallest power peak-to-valley difference is selected from the start-stop time scheme set as the final start-stop method.

[0014] As a further method, the method for clustering wind turbines based on the cut-in and cut-out wind speeds of the wind turbines and the regional wind speed distribution includes:

[0015] Obtain historical operating data and historical meteorological data of all wind turbines in the wind farm;

[0016] For each wind turbine, obtain the proportion of operating time in different cut-in wind speed and cut-out wind speed intervals, construct an operating characteristic vector based on the cut-in and cut-out wind speeds, and determine the regional wind speed distribution characteristic vector corresponding to its location;

[0017] The operation characteristic vector based on the cut-in and cut-out wind speeds and the wind speed distribution characteristic vector are merged to construct a comprehensive characteristic vector for each wind turbine;

[0018] The cosine similarity of the comprehensive feature vectors between wind turbines is calculated, and wind turbines with cosine similarity greater than 0.8 are divided into the same cluster. The remaining wind turbines after the division are divided into the cluster with the closest geographical distance.

[0019] As a further method, the method for decoupling the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency includes:

[0020] Collect historical operating data of wind turbine clusters, including output power and operating frequency;

[0021] Using the continuous wavelet transform method, the wavelet coefficients at different scales and translations are calculated to obtain the time-frequency domain characteristics, which are expressed as follows:

[0022]

[0023] Among them, a is the scale parameter, b is the translation parameter, x(t) is the signal of the output power and operating frequency of the wind turbine cluster changing with time t, and ψ is the wavelet basis function;

[0024] Taking the extracted time domain features as input variables, output power and operating frequency as output variables, and minimizing the frequency deviation as the optimization goal, the projection vector is solved by the gradient descent optimization algorithm to achieve the decoupling of output power and operating frequency from different time-frequency domain features.

[0025] As a further method, the method for constructing an inertia model of the frequency regulation capability of a wind turbine cluster includes:

[0026] The decoupled time-frequency domain operation characteristics, output power and operation frequency information are used as basic data. According to the physical characteristics of the wind turbine cluster, an inertia model of the wind turbine cluster frequency regulation capability, wind speed and output power is constructed. The expression is:

[0027] F cap (v,β,TF,J,ρ,A)=[ΔP,S p ,Δf,S f ]

[0028]

[0029] Among them, F cap (v, β, TF, J, ρ, A) represents the frequency regulation capability of the wind turbine cluster, v is the wind speed, β is the pitch angle, TF is the time-frequency domain operation characteristics, J is the moment of inertia, ρ is the air density, A is the rotor swept area, ΔP is the power frequency regulation range, S p is the power frequency modulation rate, Δf is the frequency modulation range, S f is the frequency modulation rate, v min and v max are the minimum wind speed and the maximum wind speed respectively, C p (λ,β,TF) is the power coefficient, λ is the tip speed ratio, t is the time, α is the min and α max are the minimum angular acceleration and the maximum angular acceleration respectively, α is the angular acceleration, t r is the time interval, T(β,TF) is the torque.

[0030] As a further method, the method for performing multi-objective optimization with the current operating state of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints includes:

[0031] Remove wind turbines in abnormal operating state from each wind turbine cluster, and obtain the upper limit of adjustable frequency space of wind turbines in normal operating state as the first constraint condition;

[0032] According to the inertia model of each wind turbine cluster, a power frequency regulation range constraint, a power frequency regulation rate constraint, a frequency frequency regulation range constraint and a frequency frequency regulation rate constraint of each wind turbine cluster are obtained as the second constraint condition;

[0033] The optimization objectives in the multi-objective optimization include maximizing the total power generation revenue of the wind farm, minimizing power fluctuations, and maximizing the contribution to the frequency and peak regulation of the power grid;

[0034] Among them, the expression of the total power generation income of the wind farm is:

[0035]

[0036] Where T1 is the cluster operation time, n is the number of wind turbine clusters, P i,t and P i,t-1 are the active power output of the ith cluster at time t and time t-1, C t is the electricity price at time t, α1 and α2 are the penalty coefficients for power fluctuation and deviation from rated power, ∈ is a parameter to prevent the denominator from being zero, which is set to 0.01, P i,rated is the rated power of the ith cluster;

[0037] The expression of power fluctuation is:

[0038]

[0039] Among them, β1 and β2 are the weight coefficients for adjusting the influence of power variation standard deviation and power variation high-order moment on power fluctuation, is the average active power output of all clusters at time t;

[0040] Among them, the expression of contribution to grid peak regulation is:

[0041]

[0042] Among them, γ1 and γ2 are the weight coefficients of frequency modulation capability and frequency modulation response time, respectively. i,t is the frequency modulation capability that the ith cluster can provide at time t, F req,t is the frequency regulation demand of the power grid at time t, e is a constant, Δt i,t is the delay time from the i-th cluster receiving the frequency modulation command to the actual response, τ i is the response time of the i-th cluster;

[0043] Under the condition of satisfying the first constraint and the second constraint, a non-dominated sorting genetic algorithm is used to solve the problem. If there are multiple sets of solutions that meet the conditions, the solution with the largest adjustable frequency space of the remaining wind turbines is selected as the optimal solution.

[0044] As a further method, the method of optimizing the scheduling task in real time by considering the synergy between wind turbine clusters includes:

[0045] When a wind turbine cluster has insufficient output power due to a drop in wind speed, it is marked as a wind turbine cluster that needs to be compensated, and the adjustable frequency space and location information of other wind turbine clusters are obtained;

[0046] Taking the geographical distance between other wind turbine clusters and the wind turbine cluster to be compensated as the weight, and subject to the constraints of the adjustable frequency space of the wind turbine cluster, frequency regulation tasks with output power in proportion to the weight are allocated to other wind turbine clusters.

[0047] As a further method, the method of determining the start and stop time scheme set of each wind turbine in different time periods according to the scheduling task under the condition of satisfying the power balance of each wind turbine in the cluster includes:

[0048] According to the dispatching task of the wind turbine cluster and the rated power of each wind turbine in the cluster, a power balance constraint equation is constructed, which is expressed as follows:

[0049]

[0050] Where m is the number of wind turbines in the cluster, S i,t is the start / stop state of the i-th wind turbine at time t. i,t =1, indicating operation, S i,t =0, it means stop, α i is the power correction factor of the i-th wind turbine, f i (v i,t ) is the i-th wind turbine based on wind speed v i,t The power output function, v i,t is the regional wind speed of the i-th wind turbine at time t, e is a constant, γ i is the efficiency attenuation coefficient of the i-th wind turbine, T i,t is the cumulative operating time of the i-th wind turbine unit up to time t, β ij is the influence of the jth wind turbine on the power output of the ith wind turbine, P d,t is the output power that the cluster needs to meet at time t, λ t is the transmission loss coefficient at time t;

[0051] Taking minimizing the power imbalance within the cluster in each time period as the objective function, the constraint equation is solved, where the calculation formula for the power imbalance within the cluster is:

[0052]

[0053] Among them, δt is the imbalance degree of the cluster at time t, P i,t is the actual output power of the i-th wind turbine at time t;

[0054] Obtain a set of start-up time plans for each wind turbine in different time periods.

[0055] As a further method, the method of selecting a scheme with the smallest power peak-to-valley difference from the start-stop time scheme set as the final start-stop method includes:

[0056] Simulate each start and stop time in the start and stop time scheme set to obtain the power change curve of each wind turbine in each scheme;

[0057] The control period is the time from when the wind turbine cluster receives the scheduling task to when each wind turbine in the cluster completes the corresponding output power.

[0058] During the control period, the difference between the maximum and minimum output power is calculated according to the power variation curve of the wind turbine generator set, and the scheme corresponding to the minimum difference is selected as the final start-up time.

[0059] A second aspect of the present invention provides an intelligent power dispatching system for adaptive loads, comprising:

[0060] A wind turbine clustering module is used to cluster wind turbines based on their cut-in and cut-out wind speeds and regional wind speed distribution;

[0061] Cluster frequency regulation modeling module, used to decouple the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and to build an inertial model of the frequency regulation capability of the wind turbine cluster;

[0062] A multi-module optimization allocation module is used to meet the predicted load demand of the power grid, take the current operating status of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, perform multi-objective optimization, and allocate the frequency regulation tasks corresponding to the optimal solution to each wind turbine cluster;

[0063] A cluster collaborative optimization module is used to optimize the scheduling task in real time by considering the synergy between wind turbine clusters;

[0064] A cluster start-stop plan module is used to determine the start-stop time plan set of each wind turbine in different time periods according to the scheduling task in each wind turbine cluster, on the condition of satisfying the power balance of each wind turbine in the cluster;

[0065] The optimal solution selection module is used to select the solution with the smallest power peak-to-valley difference from the start-stop time solution set as the final start-stop method.

[0066] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the method steps described in the first aspect.

[0067] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0068] (1) The present invention performs cluster division based on the cut-in and cut-out wind speeds of wind turbines and the regional wind speed distribution, thereby more reasonably integrating wind turbine resources, grouping wind turbines under similar wind speed conditions together, improving wind energy utilization efficiency, and enabling wind turbines to generate electricity efficiently under different wind speed environments;

[0069] (2) The present invention decouples the different time-frequency domain operation characteristics of each wind turbine cluster and constructs an inertial model of frequency regulation capability, thereby obtaining the frequency regulation characteristics of each cluster and using this as a constraint to perform multi-objective optimization. This enables a more scientific allocation of frequency regulation tasks, effectively improving the frequency stability of the power grid, reducing the risk of frequency deviation caused by wind power fluctuations, and ensuring safe and stable operation of the power grid.

[0070] (3) In each wind turbine cluster, the present invention determines the start and stop time plan set based on the condition of satisfying the power balance of each wind turbine in the cluster, which can avoid the situation where individual wind turbines generate excessive or insufficient power, ensure the overall power generation efficiency of the cluster, balance the workload of each wind turbine, and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The present invention is a flowchart of a method for intelligent power scheduling of an adaptive load in an embodiment of the present invention.

[0072] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0074] Reference Figure 1 As shown, the present invention provides an intelligent power scheduling method for adaptive loads, comprising:

[0075] S100 clusters wind turbines based on their cut-in and cut-out wind speeds and regional wind speed distribution;

[0076] It should be explained that cluster division is to identify groups of units with similar wind speed response characteristics, which will help to formulate more suitable control strategies and operation and maintenance plans for different clusters in the future, and improve the overall power generation efficiency and operation stability of the wind farm;

[0077] It should be understood that the cut-in wind speed refers to the minimum wind speed required for the wind turbine to start generating electricity. When the wind speed is lower than the cut-in wind speed, the wind turbine cannot generate electricity normally. The cut-out wind speed refers to the wind speed at which the wind turbine stops generating electricity due to excessive wind speed. When the wind speed exceeds the cut-out wind speed, in order to protect the wind turbine and prevent it from being damaged due to excessive wind speed, the wind turbine will automatically stop generating electricity.

[0078] In the actual evaluation, one of the wind farms is located in the coastal plain, covering an area of ​​about 30 square kilometers. There are 50 wind turbines, named 1 to 50. The historical operation data of the past year was collected, and synchronous meteorological data was obtained from three surrounding meteorological stations.

[0079] In the actual evaluation, the proportion of operating time in different cut-in and cut-out wind speed ranges was calculated. For wind turbine No. 1, its operating time in the cut-in wind speed range of 3.2-3.4m / s accounted for 9.2%, and 3.4-3.6m / s accounted for 12.5%. The regional wind speed distribution feature vector was determined in combination with the wind turbine location. The two types of feature vectors were fused, and the wind farm was divided into three clusters after cosine similarity calculation. Specifically, Cluster A contains 11 units (wind turbines No. 1-11), Cluster B has 16 units (wind turbines No. 11-27), and Cluster C has 23 units (wind turbines No. 28-50).

[0080] S200 decouples the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and builds an inertial model of the frequency regulation capability of the wind turbine cluster;

[0081] It should be explained that by decoupling the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and constructing an inertial model of the frequency regulation capability of the wind turbine cluster, the frequency regulation potential of the cluster under different working conditions can be quantified, providing accurate model support for grid frequency regulation;

[0082] In the actual evaluation, for each cluster, we collected the minute-by-minute operation data within one month, and used continuous wavelet transform, Morlet wavelet as the basis function, scale parameter a changed in the range of 0.6-8 with a step size of 0.1, and translation parameter b changed in the range of 0-60 minutes with a step size of 1 minute, and calculated the wavelet coefficients to obtain the time-frequency domain characteristics; with the time domain characteristics as input, output power and operating frequency as output, and the minimum frequency deviation as the goal, the projection vector was solved using the gradient descent optimization algorithm, and decoupling was achieved after multiple rounds of iterations. After cluster A was decoupled, the average absolute error between the output power and the original data was 0.06MW, and the average absolute error between the operating frequency was 0.02Hz;

[0083] In the actual evaluation, the inertial model of the frequency regulation capability of the wind turbine cluster is constructed based on the decoupled data. For cluster B, the average wind speed is 8.1m / s, the pitch angle is 2.9°, and the moment of inertia is 1100kg·m 2 , air density is 1.22kg / m 3 The swept area of ​​the wind wheel is 78.5m 2 The power frequency modulation range is -150kW to 160kW, the power frequency modulation rate is 40kW / s, the frequency frequency modulation range is -0.4Hz to 0.4Hz, the frequency frequency modulation rate is 0.07Hz / s, and the frequency modulation capacity of cluster B at the initial moment is 110kW.

[0084] S300 satisfies the predicted load demand of the power grid, takes the current operating state of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, performs multi-objective optimization, and allocates the frequency regulation tasks corresponding to the optimal solution to each wind turbine cluster;

[0085] In the actual evaluation, for cluster C, its No. 30 wind turbine was shut down due to a fault. The upper limit of the adjustable frequency space of the remaining normally operating wind turbines was determined as the first constraint. According to the inertia model of cluster C, the power frequency regulation range, power frequency regulation rate, frequency frequency regulation range, and frequency frequency regulation rate constraints were obtained as the second constraint. The optimization goal was to maximize the total power generation revenue of the wind farm, minimize power fluctuations, and maximize the contribution to grid frequency and peak regulation. The final dispatching task was: cluster A was assigned to increase the output power by 120kW, cluster B reduced it by 70kW, and cluster C increased it by 90kW.

[0086] S400 considers the synergy between wind turbine clusters and optimizes the scheduling task in real time;

[0087] In the actual evaluation, during one of the operations, cluster C had insufficient output power due to a sudden drop in wind speed, and the adjustable frequency space and location information of clusters A and B were obtained. The adjustable frequency space of cluster A was 180kW, and the distance from cluster C was 2.5km; the adjustable frequency space of cluster B was 160kW, and the distance was 1.8km. Tasks were allocated within the adjustable frequency space limit using geographical distance as weight, specifically: cluster A was allocated an additional frequency regulation task of 50kW, and cluster B was allocated 40kW.

[0088] S500 determines, in each wind turbine cluster, a set of start and stop time schemes for each wind turbine in different time periods according to the scheduling task, on the condition that power balance of each wind turbine in the cluster is satisfied;

[0089] It needs to be explained that, through precise start and stop time schemes, the phenomenon of wind farm abandonment can be reduced, the utilization rate of wind power can be improved, and the operating costs of wind farms can be reduced;

[0090] In the actual evaluation, for cluster A, there are 11 wind turbines in total. The constraint equation is solved with minimizing the power imbalance as the objective function, and a total of 7 sets of solutions with different start and stop time combinations are obtained.

[0091] S600 selects a scheme with the smallest power peak-to-valley difference from the start-stop time scheme set as the final start-stop method.

[0092] In the actual evaluation, for cluster A, each start-stop time combination in the scheme set was simulated, and the power change curve of the wind turbine in each scheme was recorded in detail. The average control cycle was found to be 17 hours. After one-by-one comparison, it was found that the power peak-to-valley difference of the third scheme was the smallest, only 150kW. Therefore, this scheme was determined as the final wind turbine start-stop method, specifically:

[0093] In the first dispatching period (0-17 hours): Wind turbine No. 1: Started in the first hour, quickly reached a stable operating state after startup, and the output power gradually increased to about 80% of the rated power; Wind turbine No. 2: Maintained the operating state, and its output power was stable at about 1.9MW; Wind turbine No. 3: Started in the second hour, the startup process was smooth, and the power was gradually increased after startup, reaching 70% of the rated power in the fourth hour; Wind turbine No. 4: Maintained the stopped state; Wind turbine No. 5: Maintained the operation, and the output power was stable at 2.0MW; Wind turbine No. 6: Maintained the operation, and the output power was stable at 2.2MW; Wind turbine No. 7: Started in the fourth hour, the output power was stabilized at about 1.9MW, and stopped in the 12th hour; Wind turbines No. 8, 9, and 10: Started in the tenth hour, the output power was stabilized at about 2.7MW, and stopped in the 15th hour; Wind turbine No. 11: Started in the second hour, maintained the output power stable at about 2.0MW, and increased the output power to about 2.2MW in the ninth hour, and maintained until the end of the control cycle.

[0094] In this embodiment, the method for clustering wind turbines based on the cut-in and cut-out wind speeds of the wind turbines and the regional wind speed distribution includes:

[0095] Obtain historical operating data and historical meteorological data of all wind turbines in the wind farm;

[0096] For each wind turbine, obtain the proportion of operating time in different cut-in wind speed and cut-out wind speed intervals, construct an operating characteristic vector based on the cut-in and cut-out wind speeds, and determine the regional wind speed distribution characteristic vector corresponding to its location;

[0097] The operation characteristic vector based on the cut-in and cut-out wind speeds and the wind speed distribution characteristic vector are merged to construct a comprehensive characteristic vector for each wind turbine;

[0098] The cosine similarity of the comprehensive feature vectors between wind turbines is calculated, and wind turbines with cosine similarity greater than 0.8 are divided into the same cluster. The remaining wind turbines after the division are divided into the cluster with the closest geographical distance.

[0099] In this embodiment, the method for decoupling the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and the operating frequency includes:

[0100] Collect historical operating data of wind turbine clusters, including output power and operating frequency;

[0101] Using the continuous wavelet transform method, the wavelet coefficients at different scales and translations are calculated to obtain the time-frequency domain characteristics, which are expressed as follows:

[0102]

[0103] Among them, a is the scale parameter, b is the translation parameter, x(t) is the signal of the output power and operating frequency of the wind turbine cluster changing with time t, and ψ is the wavelet basis function;

[0104] Taking the extracted time domain features as input variables, output power and operating frequency as output variables, and minimizing the frequency deviation as the optimization goal, the projection vector is solved by the gradient descent optimization algorithm to achieve the decoupling of output power and operating frequency from different time-frequency domain features.

[0105] In this embodiment, the method for constructing an inertia model of the frequency regulation capability of a wind turbine cluster includes:

[0106] The decoupled time-frequency domain operation characteristics, output power and operation frequency information are used as basic data. According to the physical characteristics of the wind turbine cluster, an inertia model of the wind turbine cluster frequency regulation capability, wind speed and output power is constructed. The expression is:

[0107] F cap (v,β,TF,J,ρ,A)=[ΔP,S p ,Δf,S f ]

[0108]

[0109] Among them, F cap (v, β, TF, J, ρ, A) represents the frequency regulation capability of the wind turbine cluster, v is the wind speed, β is the pitch angle, TF is the time-frequency domain operation characteristics, J is the moment of inertia, ρ is the air density, A is the rotor swept area, ΔP is the power frequency regulation range, S p is the power frequency modulation rate, Δf is the frequency modulation range, S f is the frequency modulation rate, v min and v max are the minimum wind speed and the maximum wind speed respectively, C p (λ,β,TF) is the power coefficient, λ is the tip speed ratio, t is the time, α is the min and α max are the minimum angular acceleration and the maximum angular acceleration respectively, α is the angular acceleration, t r is the time interval, T(β,TF) is the torque.

[0110] In this embodiment, the method for performing multi-objective optimization with the current operating state of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints includes:

[0111] Remove wind turbines in abnormal operating state from each wind turbine cluster, and obtain the upper limit of adjustable frequency space of wind turbines in normal operating state as the first constraint condition;

[0112] According to the inertia model of each wind turbine cluster, a power frequency regulation range constraint, a power frequency regulation rate constraint, a frequency frequency regulation range constraint and a frequency frequency regulation rate constraint of each wind turbine cluster are obtained as the second constraint condition;

[0113] The optimization objectives in the multi-objective optimization include maximizing the total power generation revenue of the wind farm, minimizing power fluctuations, and maximizing the contribution to the frequency and peak regulation of the power grid;

[0114] Among them, the expression of the total power generation income of the wind farm is:

[0115]

[0116] Where T1 is the cluster operation time, n is the number of wind turbine clusters, P i,t and P i,t-1 are the active power output of the ith cluster at time t and time t-1, C t is the electricity price at time t, α1 and α2 are the penalty coefficients for power fluctuation and deviation from rated power, ∈ is a parameter to prevent the denominator from being zero, which is set to 0.01, P i,rated is the rated power of the ith cluster;

[0117] The expression of power fluctuation is:

[0118]

[0119] Among them, β1 and β2 are the weight coefficients for adjusting the influence of power variation standard deviation and power variation high-order moment on power fluctuation, is the average active power output of all clusters at time t;

[0120] Among them, the expression of contribution to grid peak regulation is:

[0121]

[0122] Among them, γ1 and γ2 are the weight coefficients of frequency modulation capability and frequency modulation response time, respectively. i,t is the frequency modulation capability that the ith cluster can provide at time t, F req,t is the frequency regulation demand of the power grid at time t, e is a constant, Δt i,t is the delay time from the i-th cluster receiving the frequency modulation command to the actual response, τ i is the response time of the i-th cluster;

[0123] Under the condition of satisfying the first constraint and the second constraint, a non-dominated sorting genetic algorithm is used to solve the problem. If there are multiple sets of solutions that meet the conditions, the solution with the largest adjustable frequency space of the remaining wind turbines is selected as the optimal solution.

[0124] In this embodiment, the method for optimizing the scheduling task in real time by considering the synergy between wind turbine clusters includes:

[0125] When a wind turbine cluster has insufficient output power due to a drop in wind speed, it is marked as a wind turbine cluster that needs to be compensated, and the adjustable frequency space and location information of other wind turbine clusters are obtained;

[0126] Taking the geographical distance between other wind turbine clusters and the wind turbine cluster to be compensated as the weight, and subject to the constraints of the adjustable frequency space of the wind turbine cluster, frequency regulation tasks with output power in proportion to the weight are allocated to other wind turbine clusters.

[0127] In this embodiment, the method of determining the start and stop time scheme set of each wind turbine in different time periods according to the scheduling task under the condition of satisfying the power balance of each wind turbine in the cluster includes:

[0128] According to the dispatching task of the wind turbine cluster and the rated power of each wind turbine in the cluster, a power balance constraint equation is constructed, which is expressed as follows:

[0129]

[0130] Where m is the number of wind turbines in the cluster, S i,t is the start / stop state of the i-th wind turbine at time t. i,t =1, indicating operation, S i,t =0, it means stop, α i is the power correction factor of the i-th wind turbine, f i (v i,t ) is the i-th wind turbine based on wind speed v i,t The power output function, v i,t is the regional wind speed of the i-th wind turbine at time t, e is a constant, γ i is the efficiency attenuation coefficient of the i-th wind turbine, T i,t is the cumulative operating time of the i-th wind turbine unit up to time t, β ij is the influence of the jth wind turbine on the power output of the ith wind turbine, P d,t is the output power that the cluster needs to meet at time t, λ t is the transmission loss coefficient at time t;

[0131] Taking minimizing the power imbalance within the cluster in each time period as the objective function, the constraint equation is solved, where the calculation formula for the power imbalance within the cluster is:

[0132]

[0133] Among them, δ t is the imbalance degree of the cluster at time t, P i,t is the actual output power of the i-th wind turbine at time t;

[0134] Obtain a set of start-up time plans for each wind turbine in different time periods.

[0135] In this embodiment, the method of selecting a scheme with the smallest power peak-to-valley difference from the start-stop time scheme set as the final start-stop method includes:

[0136] Simulate each start and stop time in the start and stop time scheme set to obtain the power change curve of each wind turbine in each scheme;

[0137] The control period is the time from when the wind turbine cluster receives the scheduling task to when each wind turbine in the cluster completes the corresponding output power.

[0138] During the control period, the difference between the maximum and minimum output power is calculated according to the power variation curve of the wind turbine generator set, and the scheme corresponding to the minimum difference is selected as the final start-up time.

[0139] The second aspect of the present invention further provides an adaptive load intelligent power dispatching system, comprising:

[0140] A wind turbine clustering module is used to cluster wind turbines based on their cut-in and cut-out wind speeds and regional wind speed distribution;

[0141] Cluster frequency regulation modeling module, used to decouple the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and to build an inertial model of the frequency regulation capability of the wind turbine cluster;

[0142] A multi-module optimization allocation module is used to meet the predicted load demand of the power grid, take the current operating status of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, perform multi-objective optimization, and allocate the frequency regulation tasks corresponding to the optimal solution to each wind turbine cluster;

[0143] A cluster collaborative optimization module is used to optimize the scheduling task in real time by considering the synergy between wind turbine clusters;

[0144] A cluster start-stop plan module is used to determine the start-stop time plan set of each wind turbine in different time periods according to the scheduling task in each wind turbine cluster, on the condition of satisfying the power balance of each wind turbine in the cluster;

[0145] The optimal solution selection module is used to select the solution with the smallest power peak-to-valley difference from the start-stop time solution set as the final start-stop method.

[0146] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0147] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0148] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0149] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming an adaptive load intelligent power scheduling system at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned adaptive load intelligent power scheduling methods.

[0150] The above application Figure 1The embodiment shown discloses an adaptive load intelligent power scheduling method that can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0151] The electronic device may also perform Figure 1 A port material movement scheduling method based on multimodal pathfinding algorithm is implemented Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0152] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent power scheduling method for adaptive loads, characterized in that: The following steps are involved: Wind turbines are clustered based on their cut-in and cut-out wind speeds and regional wind speed distribution; Decouple the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and build an inertial model of the frequency regulation capability of the wind turbine cluster; Under the condition of meeting the predicted load demand of the power grid, multi-objective optimization is performed with the current operating status of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, and the frequency regulation tasks corresponding to the optimal solution are allocated to each wind turbine cluster; Considering the synergy between wind turbine clusters, the scheduling task is optimized in real time; In each wind turbine cluster, a start and stop time plan set for each wind turbine in different time periods is determined according to the scheduling task, subject to satisfying the power balance of each wind turbine in the cluster; A scheme with the smallest power peak-to-valley difference is selected from the start-stop time scheme set as the final start-stop method.

2. The method for intelligent power scheduling of an adaptive load according to claim 1, characterized in that: The method for clustering wind turbines based on the cut-in and cut-out wind speeds of the wind turbines and the regional wind speed distribution includes: Obtain historical operating data and historical meteorological data of all wind turbines in the wind farm; For each wind turbine, obtain the proportion of operating time in different cut-in wind speed and cut-out wind speed intervals, construct an operating characteristic vector based on the cut-in and cut-out wind speeds, and determine the regional wind speed distribution characteristic vector corresponding to its location; The operation characteristic vector based on the cut-in and cut-out wind speeds and the wind speed distribution characteristic vector are merged to construct a comprehensive characteristic vector for each wind turbine; The cosine similarity of the comprehensive feature vectors between wind turbines is calculated, and wind turbines with cosine similarity greater than 0.8 are divided into the same cluster. The remaining wind turbines after the division are divided into the cluster with the closest geographical distance.

3. The method for intelligent power scheduling of an adaptive load according to claim 1, characterized in that: The method for decoupling the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and the operating frequency includes: Collect historical operating data of wind turbine clusters, including output power and operating frequency; Using the continuous wavelet transform method, the wavelet coefficients at different scales and translations are calculated to obtain the time-frequency domain characteristics, which are expressed as follows: Among them, a is the scale parameter, b is the translation parameter, x(t) is the signal of the output power and operating frequency of the wind turbine cluster changing with time t, and ψ is the wavelet basis function; Taking the extracted time domain features as input variables, output power and operating frequency as output variables, and minimizing the frequency deviation as the optimization goal, the projection vector is solved by the gradient descent optimization algorithm to achieve the decoupling of output power and operating frequency from different time-frequency domain features.

4. The method for intelligent power scheduling of an adaptive load according to claim 1, characterized in that: The method for constructing an inertia model of the frequency regulation capability of a wind turbine cluster comprises: The decoupled time-frequency domain operation characteristics, output power and operation frequency information are used as basic data. According to the physical characteristics of the wind turbine cluster, an inertia model of the wind turbine cluster frequency regulation capability, wind speed and output power is constructed. The expression is: F cap (v,β,TF,J,ρ,A)=[ΔP,S p ,Δf,S f ] Among them, F cap (v, β, TF, J, ρ, A) represents the frequency regulation capability of the wind turbine cluster, v is the wind speed, β is the pitch angle, TF is the time-frequency domain operation characteristics, J is the moment of inertia, ρ is the air density, A is the rotor swept area, ΔP is the power frequency regulation range, S p is the power frequency modulation rate, Δf is the frequency modulation range, S f is the frequency modulation rate, v min and v max are the minimum wind speed and the maximum wind speed respectively, C p (λ,β,TF) is the power coefficient, λ is the tip speed ratio, t is the time, α is the min and α max are the minimum angular acceleration and the maximum angular acceleration respectively, α is the angular acceleration, t r is the time interval, T(β,TF) is the torque.

5. The method for intelligent power scheduling of adaptive load according to claim 1, characterized in that: The method for performing multi-objective optimization with the current operating state of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints includes: Remove wind turbines in abnormal operating states from each wind turbine cluster, and obtain the upper limit of adjustable frequency space of wind turbines in normal operating states as the first constraint condition; According to the inertia model of each wind turbine cluster, a power frequency regulation range constraint, a power frequency regulation rate constraint, a frequency frequency regulation range constraint and a frequency frequency regulation rate constraint of each wind turbine cluster are obtained as the second constraint condition; The optimization objectives in the multi-objective optimization include maximizing the total power generation revenue of the wind farm, minimizing power fluctuations, and maximizing the contribution to the frequency and peak regulation of the power grid; Among them, the expression of the total power generation income of the wind farm is: Where T1 is the cluster operation time, n is the number of wind turbine clusters, P i,t and P i,t-1 are the active power output of the ith cluster at time t and time t-1, C t is the electricity price at time t, α1 and α2 are the penalty coefficients for power fluctuation and deviation from rated power, ∈ is a parameter to prevent the denominator from being zero, which is set to 0.01, P i,rated is the rated power of the ith cluster; The expression of power fluctuation is: Among them, β1 and β2 are the weight coefficients for adjusting the influence of power variation standard deviation and power variation high-order moment on power fluctuation, is the average active power output of all clusters at time t; Among them, the expression of contribution to grid peak regulation is: Among them, γ1 and γ2 are the weight coefficients of frequency modulation capability and frequency modulation response time, respectively. i,t is the frequency modulation capability that the ith cluster can provide at time t, F req,t is the frequency regulation demand of the power grid at time t, e is a constant, Δt i,t is the delay time from the i-th cluster receiving the frequency modulation command to the actual response, τ i is the response time of the i-th cluster; Under the condition of satisfying the first constraint and the second constraint, a non-dominated sorting genetic algorithm is used to solve the problem. If there are multiple sets of solutions that meet the conditions, the solution with the largest adjustable frequency space of the remaining wind turbines is selected as the optimal solution.

6. The method for intelligent power scheduling of adaptive load according to claim 1, characterized in that: The method for optimizing the scheduling task in real time by considering the synergy between wind turbine clusters includes: When a wind turbine cluster has insufficient output power due to a drop in wind speed, it is marked as a wind turbine cluster that needs to be compensated, and the adjustable frequency space and location information of other wind turbine clusters are obtained; Taking the geographical distance between other wind turbine clusters and the wind turbine cluster to be compensated as the weight, and subject to the constraints of the adjustable frequency space of the wind turbine cluster, frequency regulation tasks with output power in proportion to the weight are allocated to other wind turbine clusters.

7. The method for intelligent power supply scheduling of an adaptive load according to claim 1, characterized in that: The method of determining the start and stop time scheme set of each wind turbine in different time periods according to the scheduling task under the condition of satisfying the power balance of each wind turbine in the cluster comprises: According to the dispatching task of the wind turbine cluster and the rated power of each wind turbine in the cluster, a power balance constraint equation is constructed, which is expressed as follows: Where m is the number of wind turbines in the cluster, S i,t is the start / stop state of the i-th wind turbine at time t. i,t =1, indicating operation, S i,t =0, it means stop, α i is the power correction factor of the i-th wind turbine, f i (v i,t ) is the i-th wind turbine based on wind speed v i,t The power output function, v i,t is the regional wind speed of the i-th wind turbine at time t, e is a constant, γ i is the efficiency attenuation coefficient of the i-th wind turbine, T i,t is the cumulative operating time of the i-th wind turbine unit up to time t, β ij is the influence of the jth wind turbine on the power output of the ith wind turbine, P d,t is the output power that the cluster needs to meet at time t, λ t is the transmission loss coefficient at time t; Taking minimizing the power imbalance within the cluster in each time period as the objective function, the constraint equation is solved, where the calculation formula for the power imbalance within the cluster is: Among them, δ t is the imbalance degree of the cluster at time t, P i,t is the actual output power of the i-th wind turbine at time t; Obtain a set of start-up time plans for each wind turbine in different time periods.

8. The method for intelligent power scheduling of adaptive load according to claim 1, characterized in that: The method of selecting a scheme with the smallest power peak-to-valley difference from the start-stop time scheme set as the final start-stop method includes: Simulate each start and stop time in the start and stop time scheme set to obtain the power change curve of each wind turbine in each scheme; The control period is the time from when the wind turbine cluster receives the scheduling task to when each wind turbine in the cluster completes the corresponding output power. During the control period, the difference between the maximum and minimum output power is calculated according to the power variation curve of the wind turbine generator set, and the scheme corresponding to the minimum difference is selected as the final start-up time.

9. An adaptive load intelligent power dispatching system, used to execute an adaptive load intelligent power dispatching method according to any one of claims 1 to 8, characterized in that: The system comprises: A wind turbine clustering module is used to cluster wind turbines based on their cut-in and cut-out wind speeds and regional wind speed distribution; Cluster frequency regulation modeling module, which is used to decouple the operating characteristics of each wind turbine cluster in different time and frequency domains from the output power and operating frequency, and to build an inertial model of the frequency regulation capability of the wind turbine cluster; A multi-module optimization allocation module is used to meet the predicted load demand of the power grid, take the current operating status of the wind turbine cluster and the inertia model of the frequency regulation capability as constraints, perform multi-objective optimization, and allocate the frequency regulation tasks corresponding to the optimal solution to each wind turbine cluster; A cluster collaborative optimization module is used to optimize the scheduling task in real time by considering the synergy between wind turbine clusters; A cluster start-stop plan module is used to determine the start-stop time plan set of each wind turbine in different time periods according to the scheduling task in each wind turbine cluster, subject to the condition of satisfying the power balance of each wind turbine in the cluster; The optimal solution selection module is used to select the solution with the smallest power peak-to-valley difference from the start-stop time solution set as the final start-stop method.

10. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 8.

Images