Sea fan unit maintenance method based on equipment condition-based state and fault alarm threshold value

Abstract

The invention relates to the technical field of sea power operation and maintenance, and provides a sea wind turbine generator set maintenance method based on an equipment condition-based state and a fault alarm threshold value, and the method comprises the steps: obtaining the condition-based state information of a sea wind turbine generator set; wherein the on-condition state information is equipment operation environment information of the offshore wind turbine generator in the operation process and equipment state information under the operation environment information; according to the condition-based state information, constructing a maintenance decision model based on the maintenance cost of the offshore wind turbine generator; wherein the maintenance decision model determines a maintenance strategy based on the minimum maintenance mean value of the maintenance risk indexes in unit time; and according to the maintenance decision model, determining a maintenance trigger item of the offshore wind turbine generator, and executing complete-cycle maintenance supervision of the offshore wind turbine generator corresponding to the maintenance trigger item.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance management of offshore wind power, and particularly relates to a maintenance method for offshore wind turbines based on equipment condition-based status and fault alarm thresholds. Background Art

[0002] Offshore wind turbines have the characteristics of wide distribution range, multiple management levels, and great maintenance difficulty. The marine hydrological and meteorological environment is very complex, with alternating marine climates such as monsoons and typhoons, and the erosion of seawater on wind power equipment. Coupled with water transportation and manpower limitations, the effective operation time of offshore wind power daily maintenance and management is greatly compressed. In the face of special meteorological conditions (such as heavy fog and typhoons), it will directly affect the development of offshore operation and maintenance work and directly affect the power generation efficiency.

[0003] After a fault occurs in an offshore wind turbine, the actual maintenance process can be divided into four stages:

[0004] ① The control center receives the alarm signal, and the duty personnel formulate a maintenance strategy and arrange the maintenance stage;

[0005] ② The preparation stage of maintenance resources (personnel, spare parts, transportation methods);

[0006] ③ The waiting stage for suitable weather for offshore operations;

[0007] ④ The stage of executing maintenance activities, as Figure 2 shown.

[0008] However, the existing maintenance system is aging and cannot combine multiple factors such as time, cost, and weather to reasonably arrange a maintenance plan under the operating state of offshore wind turbines.

[0009] For example, in patent CN108027908A, it proposes to construct an operation and maintenance process using a simulation platform and combine environmental parameters and Fengji performance parameters. However, for weather changes and spare parts transportation, it adopts static assumptions instead of a dynamic feedback mechanism, resulting in a large deviation between the simulation results and the actual operation and maintenance scenarios and making it difficult to support accurate decision-making.

[0010] It has the problem that the path planning of maintenance ships does not consider the non-linear impact of real-time wave height thresholds fluctuations (weather problems) on sailing time, resulting in a high possibility of maintenance delays;

[0011] Its spare parts inventory model relies on fixed threshold warnings and does not introduce random variables of supply chain delays, making it easy to have insufficient spare parts preparation;

[0012] Moreover, it mostly adopts after-sales maintenance instead of prior preparation and prior supervision, resulting in effective maintenance often being divorced from reality, and the effective maintenance time will increase due to delays, and the cost will also increase. Summary of the Invention

[0013] The present invention provides a maintenance method for an offshore wind turbine based on the equipment condition-based status and the fault alarm threshold. Through accurate condition-based status information and a maintenance decision-making model, unnecessary maintenance and over-maintenance are avoided, the maintenance cost is reduced, and the minimum maintenance mean value of the maintenance risk index per unit time is considered with the maintenance cost as the core. The maintenance risk index may include the probability of maintenance failure, the impact of maintenance on the operation of the unit, etc. Through comprehensive analysis of these factors, the optimal maintenance strategy is determined to maximize cost-effectiveness; according to the maintenance decision-making model, it is determined under what circumstances maintenance is required, that is, the maintenance trigger item. Once the trigger item is met, full-cycle maintenance supervision is immediately executed to ensure that the maintenance work is carried out according to the predetermined strategy, and to ensure the maintenance quality and the normal operation of the unit.

[0014] In a first aspect, an embodiment of the present application provides a maintenance method for an offshore wind turbine based on the equipment condition-based status and the fault alarm threshold, including:

[0015] Obtaining the condition-based status information of the offshore wind turbine; wherein, the condition-based status information is the equipment operation environment information of the offshore wind turbine during operation, and the equipment status information under the operation environment information;

[0016] According to the condition-based status information, constructing a maintenance decision-making model based on the maintenance cost of the offshore wind turbine; wherein, the maintenance decision-making model determines the maintenance strategy based on the minimum maintenance mean value of the maintenance risk index per unit time;

[0017] According to the maintenance decision-making model, determining the maintenance trigger item of the offshore wind turbine and executing full-cycle maintenance supervision of the offshore wind turbine corresponding to the maintenance trigger item.

[0018] During the implementation of the present application, by obtaining the status information of the wind power equipment and the operation environment information of the wind power equipment in real time, breaking through the traditional fixed threshold and maintenance cycle, adjusting the maintenance strategy for different condition-based statuses, reducing the risk of sudden failures, and taking the minimum maintenance mean value of the maintenance risk index per unit time as the core during the maintenance process, comprehensively considering the downtime cost, the resources consumed during the maintenance process, and the environmental weight, to achieve the lowest cost in the full cycle, and during the process of risks or failures, the maintenance trigger item and the full-cycle supervision form a closed-loop feedback to ensure the continuity and timeliness of fault warning and maintenance execution.

[0019] Combined with the first aspect, the obtaining of the condition-based status information of the offshore wind turbine further includes:

[0020] According to the condition-based status information, configuring a maintenance mode based on the fault warning threshold; wherein, the maintenance mode includes:

[0021] A component maintenance mode based on any component of the offshore wind turbine exceeding the first fault warning threshold;

[0022] Preventive maintenance mode based on the failure prediction value of an offshore wind turbine exceeding the second failure alarm threshold;

[0023] Ideal maintenance mode based on the scenario threshold corresponding to the equipment operating environment information of the offshore wind turbine in the component maintenance mode exceeding the third failure alarm threshold; wherein, there is an optimal maintenance time in the ideal maintenance mode.

[0024] During the implementation of this application, through the triggering of the maintenance mode, the triggering and falling of the three-level threshold can be achieved, covering sudden failures, environmental coupling failures, and fault scenario failures. The ideal maintenance mode combined with the dynamic calculation of the scenario threshold can average the optimal maintenance time, average maintenance cost, and fault risk. Break through the static nature of the alarm of a single threshold.

[0025] Combined with the first aspect, the maintenance decision-making model is applied to the first maintenance decision-making layer when the offshore wind turbine is in the working state and the second maintenance decision-making layer when it is in the shutdown state;

[0026] Among them, the first maintenance decision-making layer is used to receive the first request signal or the second request signal, and generate a direct response strategy for the second request signal among the maintenance strategies corresponding to the first request signal; wherein, the first request signal is a maintenance warning request signal, and the second request signal is an operation failure maintenance request signal;

[0027] The second maintenance decision-making layer is used to configure the first maintenance strategy based on the first request signal or the second maintenance strategy based on the second request signal in the first maintenance decision-making layer; wherein, the shutdown time and warning time of the offshore wind turbine in the maintenance strategy configured in the second maintenance decision-making layer follow the Weibull distribution.

[0028] During the implementation of this application, through the hierarchical processing of the first maintenance decision-making layer and the second maintenance decision-making layer, the real-time fault response can be improved, the decoupling of resource scheduling can be achieved, the maintenance efficiency can be enhanced, and the inspections of shutdown and power generation can be avoided.

[0029] Combined with the first aspect, the maintenance decision-making model is configured with maintenance optimization indicators; wherein, the maintenance optimization indicators include the shutdown time indicator per unit time and the average maintenance cost indicator per unit time;

[0030] Based on the shutdown time indicator per unit time and combined with the average maintenance cost indicator per unit time under the first maintenance strategy, the first optimized maintenance decision parameter based on the shutdown time as the main indicator is formed;

[0031] Based on the average maintenance cost indicator per unit time and combined with the shutdown time indicator per unit time under the second maintenance strategy, the second optimized maintenance decision parameter based on the average maintenance cost indicator as the main indicator is formed.

[0032] During the implementation of this application, by defining the combined weights of the downtime index and the average maintenance cost index, the optimization target priority is dynamically adjusted and optimized in the scenarios of sudden failures and planned maintenance, taking into account both power generation efficiency and economy. Incorporating downtime and maintenance costs into a unified decision-making framework can avoid resource waste or risk accumulation caused by optimizing a single index.

[0033] Combined with the first aspect, the maintenance decision-making model is also used to construct a first life cycle model based on the expected maintenance cost index and a second life cycle model based on the life cycle length index for the offshore wind turbine within the preset life cycle.

[0034] Among them, under the first life cycle model, the failure maintenance of the offshore wind turbine conforms to the Weibull distribution, and when the offshore wind turbine performs preventive maintenance, the time difference between the actual failure time and the failure warning time conforms to the overall distribution.

[0035] The second life cycle model aims to minimize the expected value of the cost per unit time within the life cycle of the offshore wind turbine and constitutes the optimization parameters for the maintenance decision-making of single components.

[0036] During the implementation of this application, the randomness of failure maintenance is modeled through the Weibull distribution, and at the same time, the "overall distribution of time difference" of preventive maintenance is introduced to avoid over-maintenance or under-maintenance and improve the reliability of the unit. With the goal of the economy of the entire life cycle, the maintenance strategy of single components is optimized to reduce the total long-term operation and maintenance cost. By dynamically correcting the preventive maintenance window, the failure of the model caused by sudden environmental changes can be avoided.

[0037] Combined with the first aspect, the maintenance trigger items of the offshore wind turbine are determined according to the maintenance decision-making model, including:

[0038] According to the maintenance decision-making model, when the offshore wind turbine is in the normal operation state, continuously monitor whether the wind speed is higher than the threshold.

[0039] When the wind speed is higher than the threshold, further determine whether the regular maintenance time has arrived.

[0040] If the regular maintenance time has not arrived, stop the machine and mark it as abnormal wind speed.

[0041] If the regular maintenance time has arrived, stop the machine and mark it as regular maintenance, generate the first maintenance trigger item, and send the "repaired" signal after completing the maintenance operation.

[0042] During the operation of the equipment, when the wind speed is not higher than the threshold, judge whether a random failure has occurred.

[0043] If a random failure occurs, stop the machine and mark it as equipment failure, generate the second maintenance trigger item, report the failure type and wait for on-site repair, and send the "repaired" signal after the repair is completed.

[0044] During the implementation of this application, through the dual judgment of the wind speed threshold and the regular maintenance time, "wind speed abnormal shutdown" and "planned maintenance shutdown" are distinguished, avoiding false triggering of unnecessary repairs in high wind speed environments and reducing power generation losses. After the repair is completed, a "repaired" signal is sent to form a "trigger - execution - feedback" closed loop to ensure real-time update of the repair status and avoid repeated shutdowns.

[0045] Combined with the first aspect, determining the maintenance trigger items of the offshore wind turbine according to the maintenance decision model further includes:

[0046] According to the maintenance decision model, a maintenance status flag unit is generated, and the operating status of the maintenance status flag unit is monitored in real time and it is judged whether it is in the rest state, normal working state or offshore operation state. Among them, in the rest state, it switches to the normal working state according to the start of the shift or the handover order; in the normal working state, it receives maintenance orders and judges whether there are orders. When there are orders, it enters the waiting for offshore / ready state;

[0047] Receiving the setout instruction makes the system enter the offshore operation state from the waiting for offshore / ready state. In the offshore operation state, equipment maintenance is carried out, and after the maintenance is completed, a back message is triggered;

[0048] The back message is fed back to the system, and after confirmation is completed, it returns to the maintenance center;

[0049] After returning to the maintenance center, maintenance information is generated and uploaded to the maintenance status flag unit, and an inspection report is automatically generated and submitted.

[0050] During the implementation of this application, through the dynamic switching of the "rest state - normal working state - offshore operation state" by the maintenance status flag unit, the full-process automated management of maintenance personnel scheduling, task execution, and feedback reporting is realized, avoiding omissions in manual handover or task delays. When in the "normal working state", it judges whether there are orders in real time and dynamically allocates maintenance team resources to avoid inefficient standby.

[0051] Combined with the first aspect, the full-cycle maintenance supervision includes:

[0052] A maintenance order receiving guidance mechanism is pre-configured to start receiving maintenance orders and generate the first process point;

[0053] Judging whether the received maintenance order is a new order according to the received maintenance order. If it is a new order, it enters the spare parts stage, generates a second process node based on the first process node, and prepares the corresponding spare parts. If it is not a new order, it directly enters the ship ready stage and generates a third process node based on the first process node;

[0054] According to the second process node, determine whether it is necessary to go to sea for maintenance. If it is necessary to go to sea for maintenance, execute the steps for going to sea for maintenance. If it is not necessary to go to sea for maintenance, directly perform equipment maintenance in the port;

[0055] After the equipment maintenance is completed, determine whether there are additional tasks. If there are additional tasks, continue with the relevant maintenance work. If there are no additional tasks, return to the maintenance center to generate an order termination node.

[0056] During the implementation of this application, through the new order determination and process node branching mechanism, distinguish between first-time maintenance and repeated maintenance tasks, reduce the risk of spare part redundancy or shortage, and improve resource utilization rate. Automatically judge the need for additional tasks after the maintenance is completed, form a "maintenance-inspection-additional" closed loop, avoid secondary sea trips caused by omitting small faults, and improve the efficiency of single sea trips.

[0057] Combined with the first aspect, the full-cycle maintenance supervision further includes:

[0058] The equipment and personnel in the operation and maintenance center are in a standby state that can respond at any time to receive maintenance dispatch orders;

[0059] Receive equipment maintenance dispatch orders from the outside through electronic systems or manual transmission, etc. The maintenance dispatch order contains the basic information of the equipment to be maintained;

[0060] Check and classify the fault types of the equipment to be maintained according to the maintenance dispatch order information, and determine the required spare parts;

[0061] Find and prepare the corresponding spare parts according to the determined fault types;

[0062] Judge the adequacy of the inventory of the prepared spare parts. If the inventory is sufficient, put the spare parts in a ready state. If the inventory is insufficient, calculate the waiting time and at the same time arrange the ship to be ready for getting the required spare parts;

[0063] After the spare parts are ready or the waiting time is determined and the ship is ready, judge the accessibility of whether the maintenance personnel can reach the location of the equipment to be maintained in time to determine the final maintenance execution plan.

[0064] During the implementation of this application, through the linkage mechanism of inventory adequacy judgment and accessibility judgment, coordinate spare parts, ship, and personnel resources in real time, avoid maintenance delays caused by shortages of a single resource, and improve the efficiency of offshore operations. When the inventory is insufficient, synchronously start the processes of obtaining spare parts and ship preparation, realize the parallelization of "resource preparation - transportation - execution", and shorten the overall maintenance cycle.

[0065] Combined with the first aspect, the implementation of the maintenance execution plan is as follows:

[0066] Monitor and obtain the current wind speed V;

[0067] Judge whether the obtained wind speed V is normal. If it is normal, proceed with subsequent calculations; if it is not normal, continue to monitor;

[0068] Through wind speed judgment, when V is less than the cut-in wind speed vin, calculate LOW_time. When vin <= V <= the cut-out wind speed vout, determine that the wind turbine is operating normally. When V is greater than vout, calculate high_time;

[0069] Update the WOT timer according to the calculation results;

[0070] Generate a WOT data report.

[0071] Loop control module: used to judge whether to continue monitoring. If continue, return to the start of the monitoring step; if not, end the entire process.

[0072] During the implementation of this application, by dividing the wind speed into cut-in wind speed, operating wind speed range, and cut-out wind speed, short downtime matching is achieved, accurately matching the maintenance window with the wind conditions, avoiding ineffective maintenance or high-risk operations. Quantify the repairable time window to provide data support for the maintenance plan.

[0073] Other features and advantages of the present invention will be described in the subsequent specification, and part of them will become obvious from the specification, or be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the written specification, claims, and drawings.

[0074] The following will further describe the technical solutions of the present invention in detail through the drawings and embodiments. Brief Description of the Drawings

[0075] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0076] Figure 1 It is a flowchart of a maintenance method for an offshore wind turbine group based on the equipment condition-based status and fault alarm threshold in an embodiment of the present invention;

[0077] Figure 2 It is a flowchart of the traditional maintenance logic in an embodiment of the present invention;

[0078] Figure 3 It is a composition diagram of the condition-based status information acquisition of an offshore wind turbine in an embodiment of the present invention;

[0079] Figure 4 It is a composition diagram of the maintenance decision-making model in an embodiment of the present invention;

[0080] Figure 5 It is the local measurement index diagram of the maintenance decision-making model in the embodiment of the present invention;

[0081] Figure 6 It is the execution diagram of the maintenance decision-making model in the embodiment of the present invention based on different life cycles of maintenance cost and maintenance events;

[0082] Figure 7 It is the operation and maintenance operation logic diagram in the embodiment of the present invention;

[0083] Figure 8 It is the business logic diagram of operation and maintenance personnel in the embodiment of the present invention;

[0084] Figure 9 It is the business logic diagram of operation and maintenance ships in the embodiment of the present invention;

[0085] Figure 10 It is the business logic diagram of spare parts in the embodiment of the present invention;

[0086] Figure 11 It is the configuration judgment logic diagram of weather downtime in the embodiment of the present invention. Detailed implementation manners

[0087] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0088] In the traditional maintenance process of offshore wind power, it is often difficult to balance the harsh offshore environment, maintenance costs, and risks during the maintenance process, resulting in very low maintenance efficiency. Moreover, due to the harsh environment where offshore wind power equipment is located, it is prone to failure. In traditional technologies, the maintenance of offshore wind power mainly relies on alarm maintenance based on fixed thresholds and periodic planned maintenance. The maintenance is not timely, resulting in a greater risk of asset loss. Moreover, periodic planned maintenance is likely to overlook potential sudden risks or equipment failures.

[0089] Embodiment 1:

[0090] Refer to Figure 1 , the structural schematic diagram of a maintenance method for offshore wind turbine groups based on equipment condition monitoring and fault alarm thresholds provided in the embodiment of the present application. It is proposed to solve the defects of traditional fixed thresholds or periodic maintenance by obtaining equipment operating environment information, such as salt spray concentration, wave impact, and equipment status information, such as gearbox vibration and blade stress, and dynamically adjust the maintenance strategy according to different working conditions to reduce the occurrence and risks caused by sudden failures.

[0091] In the actual implementation process, the operating state of an offshore wind turbine is affected by the equipment operating environment and its own state. This application collects these data as condition-based status information in real time. Specifically, equipment operating environment information such as wind speed, temperature, humidity, etc., and equipment status information such as component wear degree, motor temperature, etc. Then, the real-time operating condition of the wind turbine is determined through the equipment operating environment information, constituting the condition-based status information; the condition-based status information is the basic data input into the maintenance decision-making model in this application, which is used to distinguish the relevance between the environment and the equipment state, enhance the effect of the decision-making model, and when the condition-based status information is incorrect, collected incorrectly or cannot be collected, the maintenance decision-making model will also make mistakes synchronously.

[0092] This application takes the maintenance cost as the core and constructs the entire maintenance decision-making model based on the lowest maintenance mean value of the maintenance risk index per unit time. The maintenance risk index includes the probability of maintenance failure, the impact of maintenance on the operation of the unit, etc. Finally, according to the lowest maintenance cost, a maintenance strategy is issued, and according to the maintenance decision-making model, it is determined under what circumstances maintenance is required, that is, the maintenance trigger item.

[0093] When a certain maintenance trigger item is met, full-cycle maintenance supervision is immediately executed to ensure that the maintenance work is carried out according to the predetermined strategy, guaranteeing the maintenance quality and the normal operation of the unit.

[0094] In actual implementation, the solution of this application is based on various sensors pre-installed on the wind turbine to obtain equipment operating environment information and equipment status information in real time. For example, a wind speed sensor can measure the real-time wind speed, and a temperature sensor can monitor the temperature of the motor and key components. All data information will be transmitted in real time to the comprehensive processing center for maintenance management through remote or short-range data transmission.

[0095] The staff of the comprehensive processing center will, according to the collected condition-based status information, through a pre-constructed maintenance decision-making model based on the maintenance cost, with the lowest maintenance mean value of the maintenance risk index per unit time, generate an optimal maintenance strategy based on the specific situation under the learning of complex algorithms and historical data.

[0096] For example: When it is monitored that the temperature of a certain key component of an offshore wind turbine has risen abnormally, and at the same time the wind speed has also exceeded the normal range. It can be judged through the maintenance decision-making model that there is a potential failure risk for this component, triggering the corresponding trigger item. The operation and maintenance management team will immediately initiate a maintenance plan, deploy professional maintenance personnel to carry the corresponding spare parts to the site for maintenance.

[0097] During the maintenance process, the comprehensive processing center conducts full-cycle supervision of the entire maintenance process.

[0098] Maintenance personnel perform maintenance operations according to the predetermined maintenance strategy, while real-time feedback on the maintenance progress and relevant data is provided.

[0099] Finally, the maintenance work is successfully completed, and the wind turbine returns to normal operation.

[0100] In actual implementation, traditional technologies only monitor the gearbox temperature based on a fixed threshold, without considering the accelerating effect of salt spray concentration on bearing corrosion, resulting in sudden bearing fractures. This application is based on the fusion modeling of salt spray concentration (environmental information) and gearbox vibration spectrum (condition information), triggering maintenance in the initial stage of corrosion to avoid downtime losses. It can also solve the problem that periodic maintenance cannot cover the dynamic stress fluctuations of the blades during typhoons, resulting in the failure to detect the expansion of blade cracks in a timely manner.

[0101] Yes, this application can prevent the further deterioration of equipment failures, reduce maintenance costs, and improve the overall economic efficiency and reliability of the wind farm. This application relies on the real-time output of maintenance trigger items to form a closed loop of "monitoring - decision - execution - feedback", otherwise the maintenance effect cannot be continuously optimized.

[0102] Embodiment 2:

[0103] Refer to Figure 3 , this application provides a method for obtaining the condition-based status information of an offshore wind turbine. First, the condition-based status information of the offshore wind turbine is obtained. Then, three maintenance modes are configured according to this information: component maintenance mode, preventive maintenance mode, and ideal maintenance mode. For the component maintenance mode, it is determined whether any component of the offshore wind turbine exceeds the first fault warning threshold, and if so, component maintenance is performed. For the preventive maintenance mode, it is determined whether the fault prediction value of the offshore wind turbine exceeds the second fault alarm threshold, and if so, preventive maintenance is performed. For the ideal maintenance mode, it is determined whether the scenario threshold corresponding to the equipment operating environment information in the component maintenance mode exceeds the third fault alarm threshold, and if so, the optimal maintenance time is determined and ideal maintenance is performed.

[0104] Condition-based maintenance refers to maintenance activities based on the equipment status. According to the status of the equipment during operation and the environmental conditions it is in, the expected remaining service life or deterioration status of the equipment is predicted, and a maintenance plan is made based on this information. Its purpose is generally to improve the reliability, safety of the equipment or reduce the cost of the equipment throughout its life cycle.

[0105] Maintenance decision-making is a key link in the operation and maintenance process of offshore wind turbines. Traditional maintenance methods include corrective maintenance based on faults and preventive maintenance based on time periods, which are prone to "over-maintenance" and "under-maintenance" of components. Enterprises spend a large amount of financial resources to introduce advanced equipment condition monitoring and fault warning technologies. The ultimate goal is to guide and optimize the maintenance decision-making of offshore wind turbines so that maintenance operations are only carried out at necessary times.

[0106] In actual implementation, in an offshore wind turbine, the SCADA system monitors each component in real time. When the fault warning threshold is reached, the following decisions need to be made based on the current state of the wind turbine:

[0107] (1) Maintenance behavior decision

[0108] That is, to determine the best maintenance method. Usually, in condition-based maintenance, the allowed maintenance methods are:

[0109] ① Breakdown maintenance: Due to the complex structure of the wind power system, when a certain component fails, it may bring potential fault risks to other components. To avoid bringing fault threats to other components and serious economic losses caused by long-term downtime, when a certain component has failed, breakdown maintenance activities should be taken immediately.

[0110] ② Preventive maintenance: In the fault prediction of offshore wind power, according to the various parameters monitored in real time by the SCADA system installed on the wind turbine, when the deterioration state of the component has reached the fault warning threshold but no failure has occurred, in order to avoid serious economic losses caused by the occurrence of a failure, preventive maintenance should be carried out on this component.

[0111] (2) Optimal maintenance time

[0112] When the state of the wind turbine component has reached the fault warning threshold, the decision on the maintenance time is very important. If the maintenance time is too early, although it is beneficial to repair the component that is in the process of failing in a timely manner and avoid the failure from developing into a serious fault, due to the special operating environment of the offshore wind farm and the need to invest manpower, material resources and financial resources for each maintenance, carrying out maintenance activities on the wind turbine too early will cause a large amount of resource waste; on the contrary, if the maintenance time is too late, although a large amount of resource waste is avoided, the probability of the wind turbine component failing to a fault is increased, posing a threat of downtime to other components and the entire wind turbine, resulting in huge economic losses.

[0113] Therefore, the scientific and reasonable maintenance method and maintenance time should be determined according to the state information monitored by the equipment in real time.

[0114] Embodiment 3:

[0115] This application proposes a method for issuing the maintenance decision plan of the maintenance decision model, specifically as Figure 4As shown, this system focuses on the maintenance decision-making of offshore wind turbines. It is divided into the first maintenance decision-making layer and the second maintenance decision-making layer according to the different operating states of the wind turbines. In actual implementation, through the hierarchical processing of the working state (the first decision-making layer) and the shutdown state (the second decision-making layer), the decoupling of real-time fault response and shutdown resource scheduling is achieved, and the maintenance efficiency is improved. The first decision-making layer quickly responds to sudden faults (the second request signal), and the second decision-making layer optimizes the shutdown time based on the Weibull distribution to reduce power generation losses. The first request signal (early warning) can trigger the pre-maintenance strategy and generate a direct response strategy (such as power reduction operation) when receiving the second request signal (fault), avoiding the conflict between shutdown and power generation losses.

[0116] The main problems solved by this application are that in the prior art, it is impossible to distinguish between the working state and the shutdown state, as well as the resulting response delay and waste of shutdown time; the prior art is mainly a maintenance model at the decision-making level, based on maintenance with an exponential distribution. This application introduces a hierarchical decision-making architecture, combines the Weibull distribution time for shutdown modeling, so as to realize the allocation of maintenance resources according to the state, and the prior art does not combine dynamic priorities (early warning / fault) with the Weibull fault model;

[0117] When the offshore wind turbine is in the working state, the first maintenance decision-making layer begins to play a role. It can receive two different types of request signals, namely the first request signal (maintenance early warning request signal) and the second request signal (operation fault maintenance request signal). When receiving the first request signal, the system will generate the first maintenance strategy according to the preset rules and algorithms. This strategy is mainly to formulate countermeasures in advance for potential problems that may occur in the wind turbine, playing a role of early warning and prevention. When receiving the second request signal, it means that the wind turbine has an operation fault. At this time, the system will generate the second maintenance strategy to quickly solve the current fault problem.

[0118] In particular, in the first maintenance decision-making layer, for the maintenance strategy generated by the first request signal, a direct response strategy for the second request signal will be further generated. This is to be able to respond more quickly and accurately based on the previous early warning information when an operation fault occurs, improve the maintenance efficiency, and reduce the impact of the fault on the operation of the wind turbine.

[0119] When the offshore wind turbine is in the shutdown state, the second maintenance decision-making layer starts to work. It will receive the first maintenance strategy and the second maintenance strategy from the first maintenance decision-making layer and configure them on this basis.

[0120] In this application, the second maintenance decision layer is configured to configure maintenance strategies, that is, the downtime and warning time of the offshore wind turbine follow the Weibull distribution. The Weibull distribution is a common probability distribution and is widely used in reliability engineering and life data analysis. By making the downtime and warning time follow the Weibull distribution, the time required for maintenance can be predicted more accurately, maintenance resources can be arranged reasonably, and the scientific nature and rationality of the maintenance plan can be improved.

[0121] In this application, the strategies generated by the first maintenance decision layer and the strategies configured by the second maintenance decision layer will both be transmitted to the maintenance execution link.

[0122] Maintenance personnel perform maintenance on the offshore wind turbine according to the generated strategies, execute the tasks of maintenance operations, and then repair the wind turbine so that it can operate normally and ensure the stable power generation of the entire wind farm. Through this hierarchical decision-making and scientific strategy configuration, this application can effectively improve the maintenance efficiency and reliability of offshore wind turbines and reduce maintenance costs.

[0123] The hierarchical decision-making of this application depends on the input data source of the condition-based status information to ensure that the decision-making layer operates based on the real-time environment and equipment status. The Weibull distribution depends on the maintenance decision model to achieve the dynamic optimization of the cost-risk objective function.

[0124] In actual implementation:

[0125] The establishment of the wind turbine maintenance decision model needs to meet the following conditions:

[0126] (1) Assume that the wind turbine has two states: working state and shutdown state, and during each maintenance process (fault maintenance and preventive maintenance), the wind turbine is in the shutdown state.

[0127] (2) Assume that if an unexpected fault occurs before the wind turbine performs preventive maintenance, then fault maintenance is carried out on this component.

[0128] (3) When the operating state of the wind turbine reaches the fault warning threshold, preventive maintenance is carried out on the component to avoid greater losses.

[0129] (4) Due to the accuracy difference in the fault warning model, the fault warning may be issued before or after the fault. Therefore, assume that the time difference between the wind turbine fault and the warning follows a normal distribution; according to previous research, the failure type of the equipment follows the Weibull distribution. Therefore, assume that the fault time of the wind turbine follows the Weibull distribution.

[0130] (5) Assume that when the wind turbine is carrying out maintenance activities, it has the same downtime and downtime losses.

[0131] Example 4:

[0132] During the implementation of maintenance, as Figure 5 shown, the core lies in dynamically adjusting and optimizing the target priority in the scenarios of the first strategy for sudden failures and the second strategy for planned maintenance by defining the combined weights of the downtime index and the average maintenance cost index, taking into account both power generation efficiency and economy, combining different maintenance strategies, generating optimized maintenance decision parameters, and then determining the best maintenance plan and implementing the maintenance. The first optimization parameter gives priority to shortening the downtime and is applicable to the scenario where power generation needs to be quickly restored in case of sudden failures; the second optimization parameter gives priority to cost control and is applicable to planned maintenance or maintenance during low electricity price periods; moreover, by incorporating downtime and maintenance costs into a unified decision-making framework, it avoids resource waste (such as over-maintenance) or risk accumulation caused by optimizing a single index.

[0133] The maintenance decision model configures maintenance optimization indicators, which include the downtime index per unit time and the average maintenance cost index per unit time, and they are directly related to the cost and efficiency during the maintenance process.

[0134] When this application is implemented, under the first maintenance strategy, the downtime index per unit time is the main one, and it is combined with the average maintenance cost index per unit time. When this application is implemented, because downtime will affect production efficiency and economic benefits.

[0135] Therefore, reducing downtime will be given priority and regarded as the main optimization goal. Then the two indicators are combined to form the first optimized maintenance decision parameter based on the downtime as the main indicator. The first optimized maintenance decision parameter comprehensively considers downtime and maintenance costs.

[0136] Under the second maintenance strategy, it is the opposite situation. The average maintenance cost index per unit time is the main one, and it is combined with the downtime index per unit time. If the maintenance cost becomes the main concern, it is mainly adopted. By optimizing the average maintenance cost index, the economic cost during the maintenance process is reduced. Similarly, the two indicators are combined to form the second optimized maintenance decision parameter based on the average maintenance cost index as the main indicator.

[0137] After obtaining the first optimized maintenance decision parameter and the second optimized maintenance decision parameter, this application will also make a joint decision. The system will comprehensively consider these two parameters according to specific situations, such as the importance of the equipment and the availability of maintenance resources, to determine the final maintenance plan. Once the maintenance plan is determined, it enters the stage of implementing the maintenance and performs actual maintenance operations on the equipment.

[0138] By reasonably configuring the maintenance optimization indicators, combining different maintenance strategies, generating targeted optimized maintenance decision parameters, while ensuring the normal operation of the equipment, the maintenance cost and downtime are reduced as much as possible, and the overall operation efficiency and economic benefits of the equipment are improved.

[0139] Example 5:

[0140] This application proposes a maintenance method based on a maintenance decision-making model under two indicators of the lowest maintenance cost and the shortest maintenance time, as Figure 6 shown. Two important life cycle models are constructed within the preset life cycle, and based on this, a maintenance strategy is formulated and maintenance is executed to achieve the full-cycle supervision and operation and maintenance of offshore wind turbines.

[0141] This application models the randomness of fault maintenance through Weibull distribution, and at the same time introduces the "overall distribution of time difference" of preventive maintenance to avoid over-maintenance or under-maintenance and improve the reliability of the unit. With the goal of full-life-cycle economy, the single-component maintenance strategy is optimized to reduce the total long-term operation and maintenance cost. The dynamic distribution adapts to different maintenance types, and can perform fault maintenance and preventive maintenance. The second life cycle model is independently optimized for key components such as gearboxes and blades to avoid the problem that the traditional overall optimization model ignores local deterioration differences, and realizes "global-local" two-dimensional cost control.

[0142] The maintenance decision-making model takes the preset life cycle of the offshore wind turbine as input and constructs the first life cycle model and the second life cycle model.

[0143] The first life cycle model is based on the expected maintenance cost index, and it conforms to the Weibull distribution when the offshore wind turbine performs fault maintenance. The Weibull distribution is a probability distribution commonly used in reliability analysis and life prediction, and is used to describe the law of equipment failure occurrence. By applying the Weibull distribution, the time and probability of failure occurrence can be accurately predicted, so as to reasonably arrange maintenance resources and reduce maintenance costs.

[0144] At the same time, when performing preventive maintenance, the time difference between the actual failure time and the failure warning time conforms to the overall distribution, indicating that the situation of failure warning and actual failure occurrence can be comprehensively considered, and then maintenance preparations can be made in advance to avoid serious damage to the equipment caused by the sudden occurrence of failure.

[0145] The second life cycle model is based on the life cycle length index, which minimizes the expected value of the cost per unit time of the offshore wind turbine within the life cycle. During implementation, by comprehensively analyzing and optimizing the cost of the equipment throughout the life cycle, the optimization parameters of the single-component maintenance decision are determined, and then on the premise of ensuring the normal operation of the equipment, the maintenance cost is reduced.

[0146] Based on the information obtained from the first life cycle model and the second life cycle model, it enters the stage of formulating the maintenance strategy. In this stage, by comprehensively considering factors such as the Weibull distribution of fault maintenance, the time difference distribution of preventive maintenance, and the optimization parameters of single-component maintenance decision-making, the most suitable maintenance strategy for offshore wind turbines is formulated.

[0147] Finally, according to the formulated maintenance strategy, perform maintenance operations. In this way, the system realizes the full-cycle supervision and operation and maintenance of offshore wind turbines, ensuring that the wind power equipment can operate stably and efficiently throughout its life cycle while reducing maintenance costs.

[0148] Example 6:

[0149] This application provides an operation and maintenance job logic. As Figure 7 shown, the prior art cannot distinguish abnormal shutdowns from planned maintenance requirements in high wind speed environments, resulting in waste of maintenance resources or delayed fault response. This application distinguishes between "wind speed abnormal shutdown" and "planned maintenance shutdown" through the dual judgment of wind speed threshold and regular maintenance time, avoiding unnecessary repairs being triggered by mistake in high wind speed environments and reducing power generation losses.

[0150] In actual implementation, the wind turbine has two states: normal operation and shutdown. When the wind speed exceeds the wind speed corresponding to the rated power, the wind turbine stops running. The first maintenance trigger item is to give priority to planned maintenance when the maintenance time arrives to ensure the long-term reliability of the equipment. The second maintenance trigger item is to quickly respond to sudden faults at low wind speeds to prevent small faults from evolving into major accidents; after the maintenance is completed, a "repaired" signal is sent to form a "trigger-execute-feedback" closed loop to ensure that the maintenance status is updated in real time and avoid repeated shutdowns.

[0151] That is, when the wind speed is less than or equal to the corresponding wind speed threshold, the wind turbine will automatically return to the normal operation state.

[0152] The operation and maintenance personnel will perform regular shutdown maintenance on the wind turbine once every two weeks. After the maintenance is completed, a "repaired" message will be triggered, and the wind turbine that receives the message will also return to the normal operation state; there is a certain probability that the wind turbine will fail and stop running every month. When a certain wind turbine fails and stops running, it will automatically report the location and type of the fault of the wind turbine to the dispatching center and wait for the operation and maintenance ship and operation and maintenance personnel to come for maintenance. When the maintenance is completed and the "repaired" message is triggered, the wind turbine will return to the normal operation state.

[0153] Example 7:

[0154] This application provides an operation and maintenance personnel business logic. As Figure 8 shown, in the traditional solution, relying on manual coordination, the maintenance team did not receive the emergency work order during the typhoon gap in the rest state, resulting in blade breakage.

[0155] Through the dynamic switching of the "rest state - normal working state - offshore operation state" by the maintenance status flag unit, the present application realizes the full - process automated management of maintenance personnel scheduling, task execution, and feedback reporting, avoiding manual handover omissions or task delays; the setout instruction drives the maintenance team to transfer from the preparation state to the offshore operation state, ensuring the timeliness of task response;

[0156] The back message and automatic report generation achieve a maintenance closed - loop, reducing manual input errors and enhancing information traceability. When in the "normal working state", it continuously judges the existence of orders in real - time and dynamically allocates maintenance team resources to avoid inefficient standby (such as remaining in the preparation state instead of being forced to go to sea when there are no orders). The present application has a task trigger and supervision framework, and refines the execution process through a state machine and an instruction mechanism, making the maintenance management efficient and reliable.

[0157] The operation and maintenance personnel are divided into two states: rest and work. When the operation and maintenance personnel are in the working state, they receive a maintenance order dispatched by the maintenance center. Then the operation and maintenance personnel wait for orders to prepare to go to sea. When they receive the "setout" command, they go to sea for offshore operations. When the maintenance is completed, the "back" message is triggered, and the operation and maintenance personnel return to the maintenance center. When the operation and maintenance personnel resume their normal working state, it means that they have completed a complete offshore operation and maintenance task.

[0158] Embodiment 8:

[0159] The present application provides an operation and maintenance ship business logic, as Figure 9 shown; in view of the spare - part preparation logic in the prior art that does not distinguish between new orders and historical orders and lacks an additional task processing mechanism, resulting in resource waste or secondary voyages to sea. The present application, through the new order determination and process node branching mechanism (spare - part stage / ship preparation stage), distinguishes between first - time maintenance and repeated maintenance tasks, reduces the risk of spare - part redundancy or shortage, and improves resource utilization rate. It dynamically selects the execution path according to the maintenance type (such as whether it is necessary to go to sea) to avoid unnecessary offshore operations (such as electrical faults that can be solved in the port), reducing operation and maintenance costs and personnel safety risks. After the maintenance is completed, it automatically judges the need for additional tasks to form a "maintenance - inspection - additional" closed - loop, avoiding secondary voyages to sea caused by omitting small faults and improving the efficiency of single - voyage operations.

[0160] First, the ship is in a moored state. When a new maintenance order is received, the ship will enter the spare parts stage and determine the corresponding spare parts to be carried according to the type of fan failure. After the spare parts and the ship are ready, it will be judged whether the current meteorological conditions are suitable for going to sea: if the wave height at this time node is less than the threshold of the navigable height, the personnel will be notified to go to sea to the wind farm for maintenance. After the maintenance is completed, it will be asked whether there are any other maintenance task requests from the dispatching center. If so, continue to sail to a new location for maintenance; otherwise, the ship will return to the maintenance center. When the ship returns to the moored state again, it means that a complete offshore maintenance task has been completed.

[0161] Embodiment 9:

[0162] This application provides a spare parts business logic. As Figure 10 shown, through the linkage mechanism of inventory sufficiency judgment and reachability judgment, this application coordinates spare parts, ships, and personnel resources in real time, avoids maintenance delays caused by shortages of single resources, improves offshore operation efficiency, dynamically calculates the waiting time of spare parts, preferentially calls local inventory or remotely allocates, reduces downtime losses, combines the ship's readiness status and personnel positions, optimizes the offshore route, and ensures that the maintenance team arrives at the faulty unit in the shortest time.

[0163] When the inventory is insufficient, the spare parts acquisition and ship preparation processes are started synchronously to parallelize "resource preparation - transportation - execution" and shorten the overall maintenance cycle.

[0164] The spare parts business logic is included as part of the ship preparation. After receiving the maintenance dispatch order, the operation and maintenance center checks the type of failure and prepares spare parts according to the type of failure. If there are no or insufficient spare parts, the waiting time for spare parts needs to be calculated. If the spare parts are sufficient, the status of the spare parts is set to ready. After the spare parts and the ship are ready, the reachability is judged.

[0165] Embodiment 10:

[0166] This application is configured with a configuration judgment logic for weather downtime, as Figure 11 shown, WOT (weather downtime): The reasons caused by the weather will affect the operation of the fan. The weather downtime calculates the time when the fan does not generate electricity due to the reason of below the cut-in wind speed or above the cut-out wind speed under the condition that the fan status is normal.

[0167] The judgment logic designed in the simulation system is that according to the recorded wind speed conditions at the location of the wind farm, the cut-in wind speed and cut-out wind speed of the unit are used as the lower and upper limits of the threshold. The time when the wind speed is lower than the cut-in wind speed and the time when the wind speed is lower than the cut-out wind speed both belong to WOT.

[0168] The simulation system designs a WOT timer. According to the set cut-in wind speed and cut-out wind speed thresholds, it records the time of shutdown due to weather reasons and sums up the time to obtain the weather shutdown time WOT.

[0169] WOT = tlow + thigh

[0170] In the above formula, tlow is the time when the wind speed is lower than the cut-in wind speed vin, and thigh is the time when the wind speed is higher than the cut-out wind speed vout.

[0171] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A maintenance method for an offshore wind turbine based on the condition-based status of the device and the fault alarm threshold, characterized in that Including: Obtaining the condition-based status information of an offshore wind turbine; wherein, the condition-based status information is the equipment operation environment information of the offshore wind turbine during operation, and the equipment status information under the operation environment information; Constructing a maintenance decision model based on the maintenance cost of the offshore wind turbine according to the condition-based status information; wherein, the maintenance decision model determines the maintenance strategy based on the lowest maintenance mean value of the maintenance risk index per unit time; Determining the maintenance trigger items of the offshore wind turbine according to the maintenance decision model, and performing full-cycle maintenance supervision of the offshore wind turbine corresponding to the maintenance trigger items.

2. The maintenance method of an offshore wind turbine based on the equipment condition-based status and the fault alarm threshold as described in claim 1, wherein, The obtaining of the condition-based status information of the offshore wind turbine further includes: Configuring a maintenance mode based on a fault warning threshold according to the condition-based status information; wherein, the maintenance mode includes: A component maintenance mode based on any component of the offshore wind turbine exceeding a first fault warning threshold; A preventive maintenance mode based on the fault prediction value of the offshore wind turbine exceeding a second fault alarm threshold; An ideal maintenance mode based on the scenario threshold corresponding to the equipment operation environment information of the offshore wind turbine in the component maintenance mode exceeding a third fault alarm threshold; wherein, there is an optimal maintenance time in the ideal maintenance mode.

3. The maintenance method for an offshore wind turbine based on the condition-based status of the device and the fault alarm threshold as described in claim 1, characterized in that The maintenance decision model is applied to a first maintenance decision layer when the offshore wind turbine is in a working state and a second maintenance decision layer when the offshore wind turbine is in a shutdown state; Wherein, the first maintenance decision layer is used to receive a first request signal or a second request signal, and generate a direct response strategy for the second request signal among the maintenance strategies corresponding to the first request signal; wherein, the first request signal is a maintenance warning request signal, and the second request signal is an operation fault maintenance request signal; The second maintenance decision layer is used to configure a first maintenance strategy based on the first request signal or a second maintenance strategy based on the second request signal in the first maintenance decision layer; wherein, the shutdown time and the warning time of the offshore wind turbine in the maintenance strategy configured in the second maintenance decision layer follow a Weibull distribution.

4. The maintenance method of an offshore wind turbine based on the equipment condition-based status and fault alarm threshold according to claim 3, characterized in that, The maintenance decision model is configured with maintenance optimization indicators; wherein, the maintenance optimization indicators include a shutdown time indicator per unit time and an average maintenance cost indicator per unit time; Combining the average maintenance cost indicator per unit time based on the shutdown time indicator per unit time under the first maintenance strategy constitutes a first optimized maintenance decision parameter based on the shutdown time as the main indicator; Combining the shutdown time indicator per unit time based on the average maintenance cost indicator per unit time under the second maintenance strategy constitutes a second optimized maintenance decision parameter based on the average maintenance cost indicator as the main indicator.

5. The maintenance method of an offshore wind turbine based on the condition-based status of the device and the fault alarm threshold as described in claim 1, characterized in that, The maintenance decision model is further used to constitute a first life cycle model based on the expected maintenance cost indicator and a second life cycle model based on the life cycle length indicator during the preset life cycle of the offshore wind turbine; Wherein, under the first life cycle model, the offshore wind turbine performs fault maintenance in accordance with a Weibull distribution, and when the offshore wind turbine performs preventive maintenance, the time difference between the actual occurrence time of the fault and the fault warning time conforms to an overall distribution; The second life cycle model aims to minimize the expected value of the cost per unit time within the life cycle of the offshore wind turbine, and constitutes the maintenance decision optimization parameters of a single component.

6. The maintenance method of an offshore wind turbine based on the condition-based status of the device and the fault alarm threshold according to claim 1, characterized in that, Determining the maintenance trigger items of the offshore wind turbine according to the maintenance decision model includes: According to the maintenance decision model, when the offshore wind turbine is in normal operation, continuously monitor whether the wind speed is higher than the threshold; When the wind speed is higher than the threshold, further determine whether the regular maintenance time has arrived; If the regular maintenance time has not arrived, stop the machine and mark it as abnormal wind speed; If the regular maintenance time has arrived, stop the machine and mark it as regular maintenance, generate the first maintenance trigger item, and send a "repaired" signal after completing the maintenance operation; During the operation of the equipment, when the wind speed is not higher than the threshold, determine whether a random failure has occurred; If a random failure occurs, stop the machine and mark it as equipment failure, generate the second maintenance trigger item, report the failure type and wait for on-site repair, and send a "repaired" signal after the repair is completed.

7. The maintenance method of an offshore wind turbine based on the equipment condition-based status and fault alarm threshold as described in claim 1, characterized in that, Determining the maintenance trigger items of the offshore wind turbine according to the maintenance decision model further includes: According to the maintenance decision model, generate a maintenance status flag unit, and continuously monitor the operation status of the maintenance status flag unit and determine whether it is in a rest state, normal working state or offshore operation state. Among them, in the rest state, switch to the normal working state according to the shift start or handover instruction; in the normal working state, receive a maintenance order and determine whether there is an order. When there is an order, enter the waiting for offshore / ready state; Receive the setout instruction to make the system enter the offshore operation state from the waiting for offshore / ready state. In the offshore operation state, implement equipment maintenance, and trigger a back message after the maintenance is completed; Feed back the back message to the system, and return to the maintenance center after confirmation; After returning to the maintenance center, generate maintenance information and upload it to the maintenance status flag unit, and automatically generate and submit an inspection report.

8. The maintenance method of an offshore wind turbine based on the condition-based status of the device and the fault alarm threshold according to claim 1, characterized in that The full-cycle maintenance supervision includes: Pre-configure a maintenance order receiving guidance mechanism, start receiving maintenance orders, and generate the first process point; Judge whether the received maintenance order is a new order according to the received maintenance order. If it is a new order, enter the spare parts stage, generate a second process node based on the first process node, and prepare the corresponding spare parts. If it is not a new order, directly enter the ship ready stage and generate a third process node based on the first process node; According to the second process node, judge whether it is necessary to go to sea for maintenance. If it is necessary to go to sea for maintenance, execute the offshore maintenance steps. If it is not necessary to go to sea for maintenance, directly perform equipment maintenance in the port; After completing the equipment maintenance, judge whether there is an additional task. If there is an additional task, continue with the relevant maintenance work. If there is no additional task, return to the maintenance center and generate an order termination node.

9. The maintenance method of an offshore wind turbine based on the equipment condition-based status and the fault alarm threshold as described in claim 1, characterized in that The full-cycle maintenance supervision further includes: The equipment and personnel in the operation and maintenance center are in a standby state that can respond at any time to receive maintenance dispatch orders; Receive equipment maintenance dispatch orders from the outside through electronic systems or manual transmission, etc. The maintenance dispatch orders include the basic information of the equipment to be repaired; Check and classify the failure types of the equipment to be repaired according to the maintenance dispatch order information, and determine the required spare parts; Search for and prepare the corresponding spare parts according to the determined failure types; Judge the inventory adequacy of the prepared spare parts. If the inventory is sufficient, place the spare parts in the ready state; if the inventory is insufficient, calculate the waiting time and arrange for the ship to be ready for maintenance to obtain the required spare parts. After the spare parts are ready or the waiting time is determined and the ship is ready for maintenance, judge the accessibility of the maintenance personnel to reach the location of the maintenance equipment in time to determine the final maintenance execution plan.

10. A maintenance method for an offshore wind turbine based on the equipment condition-based status and fault alarm threshold as described in claim 9, characterized in that, The implementation of the maintenance execution plan is carried out through the following steps: Monitor and obtain the current wind speed V. Judge whether the obtained wind speed V is normal. If it is normal, proceed with subsequent calculations; if it is not normal, continue to monitor. Through wind speed judgment, when V is less than the cut-in wind speed vin, calculate LOW_time; when vin <= V <= the cut-out wind speed vout, determine that the fan is operating normally; when V is greater than vout, calculate high_time. Update the WOT timer according to the calculation results. Generate a WOT data report. Loop control module: used to judge whether to continue monitoring. If continue, return to the start monitoring step; if not continue, end the entire process.

Images