Offshore wind plant operation and maintenance scheme optimization method based on optimal operation and maintenance time window

Abstract

According to the offshore wind power plant operation and maintenance scheme optimization method based on the optimal operation and maintenance time window, a reliability function is established by introducing position parameter improved Weibull distribution, a discretized black-wing optimization algorithm and an operation and maintenance scheduling model with economic cost minimization considered are applied, the maximum cargo capacity of an operation and maintenance ship is considered, and the optimal operation and maintenance time window is obtained. The optimal operation and maintenance ship scheduling and navigation path is obtained, and the problem that the operation and maintenance time cost and the operation and maintenance economic cost are increased in the prior art is solved.

Description

Technical Field

[0001] The invention relates to an offshore wind farm operation and maintenance scheme optimization method based on an optimal operation and maintenance time window, and belongs to the technical field of offshore wind power. Background Art

[0002] As an important part of clean energy, the development of offshore wind farm operation and maintenance technology is of great significance to improving energy efficiency and reducing operating costs.

[0003] The existing wind turbine fault operation and maintenance solution (see Chinese Patent Publication No. CN116664105A), although taking into account the operation and maintenance costs, does not take into account the maximum cargo capacity of the operation and maintenance ship, and cannot obtain the optimal operation and maintenance ship scheduling and navigation path, resulting in increased time and economic costs of operation and maintenance. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an offshore wind farm operation and maintenance solution optimization method based on an optimal operation and maintenance time window.

[0005] The present invention is achieved through the following technical solutions.

[0006] The present invention provides an offshore wind farm operation and maintenance scheme optimization method based on an optimal operation and maintenance time window, comprising: establishing a reliability function by introducing a position parameter to improve the Weibull distribution, applying a discretized black kite optimization algorithm and an operation and maintenance scheduling model that considers economic cost minimization, and completing the offshore wind farm operation and maintenance scheme optimization method process.

[0007] The method process comprises:

[0008] Step 1: Obtain real-time operating status information, historical fault data, and hydrological and meteorological parameter information near the offshore wind farm area.

[0009] Step 2: Establish reliability function and failure rate function based on the acquired historical failure data, and calculate the optimal operation and maintenance time window for different components of each wind turbine in the wind farm.

[0010] Step 3: Determine the operation and maintenance tasks according to the distribution of faulty wind turbines in the wind farm, the type of faulty components, and the calculated optimal operation and maintenance time window.

[0011] Step 4: Considering the maximum cargo capacity constraint of the operation and maintenance ship, the discrete black kite optimization algorithm is used to solve the operation and maintenance plan with the lowest cost. The operation and maintenance ship leasing, operation and maintenance personnel deployment and operation and maintenance path planning of the operation and maintenance plan are optimal.

[0012] The black-winged kite optimization algorithm includes four stages: population initialization stage, attack hunting stage, and seasonal migration stage.

[0013] The mathematical model of the operation and maintenance solution with the minimum cost is:

[0014]

[0015] The constraints of the operation and maintenance solution with the minimum cost are:

[0016]

[0017]

[0018]

[0019] The beneficial effects of the present invention are as follows: by introducing a position parameter to improve the Weibull distribution to establish a reliability function, applying a discretized Black Kite optimization algorithm and an operation and maintenance scheduling model that considers economic cost minimization, taking into account the maximum cargo capacity of the operation and maintenance ship, the optimal operation and maintenance ship scheduling and navigation path are obtained, thereby solving the problem of the prior art increasing the time cost and economic cost of operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is the overall flow chart of the method of the present invention;

[0021] Figure 2 It is a relationship diagram of component reliability, operation and maintenance time window, and opportunity maintenance interval of the present invention;

[0022] Figure 3 It is a distribution diagram of path decoding and encoding of the present invention;

[0023] Figure 4 It is a flow chart of the black-winged kite optimization algorithm of the present invention;

[0024] Figure 5 This is a schematic diagram of the distribution of offshore wind farms in the present invention;

[0025] Figure 6 It is the optimal path measurement cost change diagram comparing BKA, GA and ACO of the present invention;

[0026] Figure 7 It is the optimal operation and maintenance path map found by the BKA of the present invention; DETAILED DESCRIPTION

[0027] The technical solution of the present invention is further described below, but the scope of protection claimed is not limited to the description.

[0028] like Figures 1 to 7 shown.

[0029] The present application provides an offshore wind farm operation and maintenance scheme optimization method based on an optimal operation and maintenance time window, comprising the following steps. Figure 1 .

[0030] Step 1: Obtain real-time operating status information, historical fault data, and hydrological and meteorological parameter information near the offshore wind farm area.

[0031] Step 2: Establish reliability function and failure rate function based on the acquired historical failure data, and calculate the optimal operation and maintenance time window for different components of each wind turbine in the wind farm.

[0032] Reference Figure 2 The step 2 is specifically divided into eight sub-steps S2.1 to S2.8:

[0033] S2.1: Analyze the collected historical failure data, including the time of failure, component operating environment and usage, and identify the main failure modes and causes;

[0034] S2.2: Define three descriptive parameters:

[0035] Shape parameter (β): determines the trend of failure rate over time and reflects the shape of the failure distribution; helps identify whether failures are mainly caused by random factors or wear, thereby guiding the formulation of preventive maintenance strategies.

[0036] Scale parameter (η): is related to the failure rate and the service life of the component. It can be used to predict the average life or failure time of the component and is usually related to the statistical characteristics of the failure data.

[0037] Location parameter (γ): This parameter allows the model to account for shifts in the failure time distribution, since many systems may not fail for some time after they start running. This parameter can be used to adjust the model to accommodate data that has no failures during the initial time period.

[0038] S2.3: Count historical failure data and use the maximum likelihood estimation method to make more accurate estimates of shape parameters, scale parameters, and location parameters.

[0039] S2.4: Establish the reliability function R(t) and the failure rate function λ(t): The reliability function R(t) represents the The probability that the component still has no failure when , for the three-parameter Weibull distribution, the reliability function can be expressed as:

[0040]

[0041] The failure rate function λ(t) describes the Given the probability of failure without failure, the failure rate function of the three-parameter Weibull distribution is:

[0042]

[0043] S2.5: Based on the reliability function R(t), select a preventive maintenance point T p , which corresponds to the component reliability dropping to a predetermined low value R p The time point of time;

[0044] S2.6: Select an opportunity maintenance starting point T 0 , which corresponds to the reliability of the component being higher than the reliability of the preventive maintenance point R 0 time point;

[0045] S2.7: Analyze the failure rate function λ(t), identify the time point when failures increase significantly, and determine the opportunity maintenance interval ΔT, which is expressed as [T 0 ,T p ], where T 0 is the earliest time point for opportunity maintenance, and T p It is the last time point for preventive maintenance;

[0046] S2.8: Consider two situations and determine the optimal operation and maintenance time window: ① When the operation and maintenance ship departs from the port, the corresponding maintenance time T of the parts c In the opportunity maintenance interval, that is, T c ∈[T o ,T p ], then the operation and maintenance time window is T w =[0,T p -T c ], then the maximum time that the maintenance ship can spend on the journey is T p -Tc, otherwise it will cause downtime and other failures;

[0047] ②When the maintenance ship departs from the port, the corresponding maintenance time of the parts is T c Not within the opportunity maintenance interval, that is Then the operation and maintenance time window is T w =[T 0 -T c ,T p -T c ].

[0048] From the above analysis, it can be seen that the method proposed in the present invention carefully analyzes the historical failure data, uses the three-parameter Weibull distribution to establish the reliability function and the failure rate function, accurately estimates the shape parameters, scale parameters and location parameters, and then determines the preventive maintenance points and the starting points of the opportunistic maintenance.

[0049] By analyzing the failure rate function, the time point when the failure rate increases significantly is identified, the opportunity maintenance interval is determined, and based on this, the optimal operation and maintenance time window is determined.

[0050] This method not only takes into account the operating environment and usage of components, but also fully considers the scheduling and time cost of operation and maintenance vessels, thus providing a more accurate, economical and efficient operation and maintenance solution for offshore wind farms.

[0051] Through this method, the operation and maintenance efficiency of offshore wind farms can be significantly improved, the operation and maintenance costs can be reduced, the reliability and service life of wind power equipment can be improved, and ultimately the economic cost of offshore wind farm operation and maintenance can be minimized and the operation and maintenance efficiency can be maximized.

[0052] Step 3: Determine the operation and maintenance tasks according to the distribution of faulty wind turbines in the wind farm, the type of faulty components, and the calculated optimal operation and maintenance time window.

[0053] Step 4: Considering the maximum cargo capacity constraint of the operation and maintenance ship, the discrete black kite optimization algorithm is used to solve the operation and maintenance plan with the lowest cost. The operation and maintenance ship leasing, operation and maintenance personnel deployment and operation and maintenance path planning of the operation and maintenance plan are optimal.

[0054] Reference Figure 3 , the discrete method of the discrete black kite optimization algorithm in step 4 has the following characteristics: the present application adopts a discretized integer encoding mechanism to make the leader position of the black kite correspond to the solution. Assuming that the number of wind turbines that need to be maintained in a certain offshore wind farm is n, and the number of operation and maintenance ships that need to be rented when an operation and maintenance task is generated is m, the length of each solution x is n+m-1.

[0055] Let 0 represent the port number, and the other integers greater than 0 are the wind turbine numbers. Then each operation and maintenance ship departs from port 0 and returns to the port after completing the corresponding number of operation and maintenance tasks.

[0056] Assuming that the set of wind turbine numbers to be repaired is [1, 2, 3, 4, 5, 6, 7, 8, 9, 10], and the set of maintenance ship shift numbers is [11, 12, 13], the encoding and decoding of the maintenance ship's navigation path is as follows: Figure 2 shown.

[0057] The operation and maintenance path is encoded as a sequence that includes the unit number and the operation and maintenance ship shift number and ends with the operation and maintenance ship shift number; the sequence is decoded into three operation and maintenance paths: operation and maintenance ship No. 11 maintains units 1, 2, and 3, operation and maintenance ship No. 12 maintains units 4, 5, 6, and 7, and operation and maintenance ship No. 13 maintains units 8, 9, and 10.

[0058] Reference Figure 4 , the black kite optimization algorithm in step 4 includes the following features:

[0059] The black kite optimization algorithm mainly includes four stages: population initialization stage, attack hunting stage, and seasonal migration stage;

[0060] (1) Population initialization stage. The Black Kite Optimization Algorithm (BKA) belongs to the population-based primitive heuristic method, in which each black kite is considered to be a member of the algorithm population, and its position in the search space determines the value of the decision variable. Therefore, in BKA, the position of the black kite represents a candidate solution to the objective function. In the initial position stage of BKA, the position of each individual is randomly initialized in the search space by formula (3).

[0061] x ij = l j +rand()×(u j -l j ) i=1,2,K,n; j=1,2,K,m (3)

[0062] Where xij represents the j-dimensional position of the ith black-winged kite, lj and uj are the lower and upper limits respectively, and rand() represents a random number between 0 and 1.

[0063] The optimization of BKA starts from the population of candidate solutions, which is determined by the population position matrix in equation (4). Each row of the matrix represents a candidate solution, and each column of the matrix represents the solution of each dimension of the problem variable. The currently obtained optimal solution is approximately regarded as the optimal solution of each iteration.

[0064]

[0065] In the formula, X represents the black kite group, Xi represents the i-th black kite, n is the number of individuals in the black kite population, and m is the dimension of the problem variable of the objective function. Since each black kite represents a candidate solution to the optimization problem, the objective function can be evaluated based on the value proposed by each black kite for the problem variable, and then the obtained objective function value can be compiled into a vector using formula (5).

[0066]

[0067] At initialization, BKA selects the individual with the smallest fitness as the leader X in the initial population b , the leader X b It is considered to be the optimal location for black-winged kites, with the most abundant food resources. The mathematical model is shown in formula (6) and formula (7):

[0068] f best =min(f(X i )) (6)

[0069] X b=X(find(f best = = f(X i ))) (7)

[0070] (2) During the attack hunting phase, the black kite adjusts its wings and tail angle according to the wind speed during flight, hovers quietly to observe its prey, and then quickly dives to attack. This strategy includes different attack behaviors for local exploration.

[0071] By simulating the hovering observation and swooping attack behaviors of black-winged kites, the local search space can be scanned and explored in detail, thereby improving the performance of the black-winged kite algorithm (BKA) in solving optimization problems, especially the ability to efficiently find the optimal solution or approximate optimal solution in the local area. The mathematical model of the black-winged kite hovering and observing in the local space and then quickly diving to attack the prey is shown in formula (8):

[0072]

[0073] In the formula, and represents the position of the i-th black-winged kite in the j-th dimension at the t-th and t+1-th iterations, r 1 is a random number in the interval [0,1], p is a constant value of 0.9, T is the maximum number of iterations, and t is the number of iterations so far.

[0074] (3) During the seasonal migration phase, if the fitness value of the current population is less than that of the random population, the leader will give up leadership and join the migrating population, indicating that it is not suitable to lead the population forward.

[0075] On the contrary, if the fitness value of the current population is greater than the fitness value of the random population, the population will be guided to the destination. This strategy can dynamically select excellent leaders to ensure the success of migration. The mathematical model of the migration behavior of black-winged kites is shown in formula (8) and formula (9):

[0076]

[0077] In the formula, represents the leader of the j-th black-winged kite as of the t-th iteration, f i represents the fitness value of any black-winged kite at the current position of the jth dimension in the tth iteration, f ri represents the fitness value of any black-winged kite at any position in the jth dimension in the tth iteration;

[0078] r 2 is any random number in the interval [0,1], and They represent the position of the i-th black kite in the j-th dimension in the t-th and t+1-th iteration steps respectively. C(0,1) is the Cauchy mutation, which is defined as shown in formula (10):

[0079]

[0080] Where μ is the location parameter, which determines the center position of the distribution, that is, the peak position; δ is the scale parameter, which controls the width of the distribution. The larger the scale parameter, the more concentrated the data distribution.

[0081] Further, refer to Figure 5 , the specific embodiments of this application are as follows:

[0082] There are 80 wind turbines in an offshore wind farm. Referring to the distribution of wind turbines in Longyuan offshore wind farm, assuming that the wind turbine numbers and layout are as follows: Figure 5 As shown, the operation and maintenance port is 0, the orange triangle represents that the wind turbine needs maintenance, and the white one does not.

[0083] Considering the four components of wind turbines, namely the main shaft, gearbox, generator and hydraulic system, which are very important for power generation, if they are not effectively maintained in time, it will have a significant impact on power generation production.

[0084] Therefore, Weibull fitting is performed according to step 2 to calculate the reliability function R parameters and operation and maintenance time windows of the four components, as shown in Table 1.

[0085] Table 1 Reliability function R parameters and operation and maintenance time window parameters of four components

[0086]

[0087] In order to accurately describe the optimization method of the offshore wind farm with the best operation and maintenance time window, the parameter definitions shown in Table 2 are proposed:

[0088] Table 2 Parameter definition table

[0089]

[0090] With the goal of minimizing the cumulative sum of multiple costs including ship rental costs, time window violation penalty costs, etc., the following mathematical model can be proposed:

[0091]

[0092] The constraints are:

[0093]

[0094] Wherein, formula (12) is the objective function considering the total cost of ship navigation, the fixed rental cost of each ship, and the penalty cost of time window violation; formula (13) is the penalty cost of time window violation, p 1 is the penalty cost coefficient for each over-maintenance, p 2 is the penalty cost coefficient for each delayed maintenance; Formula (14) is that the number of parts required for the wind turbines on the maintenance path of each maintenance ship does not exceed the total deadweight tonnage of the maintenance ship; Formula (15) indicates that the maintenance of any unit i can be completed with only one maintenance ship; Formulas (16) and (17) indicate that the path from unit i to unit j of any maintenance ship k is irreversible and only once; Formulas (18) and (19) indicate that any maintenance ship k starts its maintenance task from the port and returns to the port after completion; Formula (20) indicates that the maintenance start time of unit j is later than the sum of the maintenance start time, maintenance time and navigation time of unit i; Formula (21) indicates the maintenance time window of any unit i.

[0095] Further references Figure 1 , the discretized black kite optimization algorithm can be used to solve Figure 7 The wind power operation and maintenance path shown in the figure is shown in the figure. In the process of finding the operation and maintenance path with the minimum cost consumption, the total cost changes with the number of iterations as shown in the figure. Figure 6 As shown in Table 3, two other algorithms are added for comparison. The operation and maintenance path solution with the lowest cost found by Black Winged Kite is shown in Table 3.

[0096] Table 3. Parameters of the minimum consumption operation and maintenance path obtained by the black kite optimization algorithm

[0097]

[0098] In summary, the offshore wind farm operation and maintenance scheme optimization method proposed in the present invention establishes a reliability function by introducing a position parameter to improve the Weibull distribution, applies a discretized Black Kite optimization algorithm and an operation and maintenance scheduling model that considers the minimization of economic costs, considers the maximum cargo capacity of the operation and maintenance ship, obtains the optimal operation and maintenance ship scheduling and navigation path, and solves the problem of increasing the time cost and economic cost of operation and maintenance in the prior art.

[0099] By introducing location parameters to improve the reliability function established by Weibull distribution, the prediction accuracy of wind turbine component failure time is improved, and the optimal operation and maintenance time window can be determined more accurately. Using the discretized Black-winged kite optimization algorithm (BKA), the lowest-cost operation and maintenance route can be generated based on comprehensive consideration of economic factors such as over-maintenance and under-maintenance penalty costs and ship leasing, effectively reducing the operation and maintenance costs. When a wind turbine fails, the model can quickly calculate the optimal operation and maintenance ship scheduling and navigation path under the maximum cargo capacity limit of the operation and maintenance ship.

Claims

1. A method for optimizing an offshore wind farm operation and maintenance plan based on an optimal operation and maintenance time window, characterized in that: include: By introducing location parameters to improve the Weibull distribution and establish a reliability function, the discretized Black Kite optimization algorithm and the operation and maintenance scheduling model considering economic cost minimization are applied to complete the optimization method process of offshore wind farm operation and maintenance solutions.

2. The offshore wind farm operation and maintenance scheme optimization method based on the optimal operation and maintenance time window according to claim 1 is characterized in that: The method process comprises: Step 1: Obtain real-time operating status information of units in the offshore wind farm, historical fault data, and hydrological and meteorological parameter information near the offshore wind farm area; Step 2: Establish a reliability function and a failure rate function based on the acquired historical failure data, and calculate the optimal operation and maintenance time window for different components of each wind turbine in the wind farm; Step 3: Determine the operation and maintenance tasks according to the distribution of the faulty wind turbines in the wind farm, the type of the faulty components, and the calculated optimal operation and maintenance time window; Step 4: Considering the maximum cargo capacity constraint of the operation and maintenance ship, the discrete black kite optimization algorithm is used to solve the operation and maintenance plan with the lowest cost. The operation and maintenance ship leasing, operation and maintenance personnel deployment and operation and maintenance path planning of the operation and maintenance plan are optimal.

3. The method for optimizing the operation and maintenance scheme of an offshore wind farm based on the optimal operation and maintenance time window according to claim 2, characterized in that: The black-winged kite optimization algorithm includes four stages: population initialization stage, attack hunting stage, and seasonal migration stage.

4. The method for optimizing the operation and maintenance scheme of an offshore wind farm based on the optimal operation and maintenance time window according to claim 3, characterized in that: The mathematical model of the operation and maintenance solution with the minimum cost is:

5. The method for optimizing the operation and maintenance scheme of an offshore wind farm based on the optimal operation and maintenance time window according to claim 3, characterized in that: The constraints of the operation and maintenance solution with the minimum cost are:

Images