Method for identifying and prompting annual maintenance plan risk of hydropower station

Abstract

A hydropower station annual maintenance plan risk identification prompting method comprises the steps of firstly obtaining detailed data such as hydropower station equipment maintenance requirements, a power grid maintenance plan and a reservoir application plan, and extracting key information such as time nodes, key period operation requirements and unit power generation capacity; models such as overhaul time overlapping judgment, incoming and outgoing line overhaul and reservoir dispatching conflict judgment, sending-out capacity judgment, repeated overhaul judgment and operation risk judgment are established; and identifying a plurality of potential risks, such as overlapping bus maintenance time of a hydropower station and a power grid converter station, shutdown of a unit caused by maintenance of incoming line equipment, insufficient sending capacity, repeated maintenance of a bus line, power grid operation risk in a specific operation mode and the like. For various risks, corresponding optimization algorithms are adopted to adjust maintenance time and operation modes and redistribute maintenance resources, it is ensured that an annual maintenance plan of the hydropower station is scientific and reasonable, the risks are reduced, it is ensured that the maximum power generation capacity of a plant station meets a reservoir application plan, and the safety and stability of hydropower station operation are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of the overhaul plan design of hydropower stations, and particularly relates to a method for risk identification and prompt of the annual overhaul plan of a hydropower station. Background Art

[0002] Affected by various random factors such as wind speed and light intensity, the output of wind power and photovoltaic power is unstable, which puts forward new requirements for the energy storage of the reservoir of the hydropower station and greatly reduces the equipment overhaul space. At the same time, in recent years, ensuring power supply during cold snaps and high temperatures has brought new challenges to the overhaul work of hydropower stations. How to formulate a reasonable overhaul plan for hydropower stations, avoid various safety risks brought by the overhaul work of hydropower stations, and create conditions for the consumption of new energy is of great significance.

[0003] In the prior art, the risks of the overhaul plan mainly rely on manual analysis for search and identification, which is greatly affected by the experience level and work attitude of personnel. The overhaul risk analysis mainly focuses on unreasonable arrangements such as time arrangement, personnel allocation, and resource allocation, and violations of regulations during the on-site overhaul work. There are mainly two drawbacks in the prior art. The first is that it is overly affected by human factors; the second is that the risk analysis focuses on management and short-term risks within the plant, and the risk analysis of annual reservoir operation and power grid operation is weak. Summary of the Invention

[0004] The technical problem to be solved by the invention is to provide a method for risk identification and prompt of the annual overhaul plan of a hydropower station, which can quickly identify the possible risks of the plant overhaul plan, prompt the corresponding level of risks, and obtain a safe, reasonable overhaul plan that meets the reservoir operation; by using this method, the unreasonable overhaul plan arrangement caused by the intricate grid relationship and the large number of in-station equipment is effectively avoided, which not only improves the efficiency but also greatly strengthens the safety construction of the power system.

[0005] To solve the above technical problems, the technical solution adopted by the invention is: A method for risk identification and prompt of the annual overhaul plan of a hydropower station, the method comprising: Risk grading: According to the risks from large to small, they are first-level risks, second-level risks, and third-level risks in sequence. The first-level risk is the overhaul work that is absolutely prohibited during the operation of the power grid, and this operation mode will cause accidents in the plant and the power grid. The second-level risk is that the constraints generated by the overhaul work cause the maximum power generation capacity of the plant to be less than the reservoir operation plan, or violate the implicit management regulations or work habits of the power grid, and has a certain degree of flexibility. The third-level risk is the overhaul work that is not very reasonable and will not cause other impacts under the current circumstances.

[0006] S1. Obtain the detailed data of the equipment overhaul requirements and the power grid overhaul plan of the hydropower station, and extract the time node information; S1.1 Obtain the maintenance data of hydropower station equipment and the grid maintenance plan, and extract the maintenance time nodes and equipment status information; S1.2 Judge the maintenance priority and maintenance type according to the maintenance time nodes and equipment status; S1.3 Determine the maintenance duration and maintenance location by using the preset maintenance cycle and grid load data; S1.4 Match the maintenance personnel and maintenance tool information through the maintenance priority and maintenance type; S1.5 If the maintenance duration exceeds the preset threshold, adjust the grid load data and maintenance plan; S1.6 Update the maintenance time nodes and maintenance priority according to the adjusted maintenance plan; S1.7 Obtain the updated maintenance plan data and generate the final maintenance arrangement information.

[0007] S2. According to the time node information, establish a maintenance time overlap judgment model to identify the time overlap risk between the hydropower station bus maintenance and the grid converter station bus maintenance; S2.1 Obtain the maintenance time points of the hydropower station bus and the grid converter station bus, generate two independent time lines, and mark the start and end times of the maintenance period respectively; S2.2 Judge whether there is an intersection between the two time lines through the preset model. If there is an intersection, determine that the maintenance periods overlap; S2.3 For the overlapping maintenance periods, extract the specific time periods on the time axis and mark them as risk points; S2.4 Generate an overlap risk report according to the time range of the risk points, including the maintenance item information of the hydropower station and the converter station; S2.5 Use machine learning algorithms to analyze the historical maintenance data and establish a time point prediction model for optimizing the maintenance time arrangement; S2.6 Adjust the maintenance time points of the hydropower station and the converter station through the time point prediction model to reduce the overlap risk; S2.7 Regenerate the time line according to the adjusted time points, and perform the overlap risk judgment again to obtain the final optimization result.

[0008] S3. If there is a time overlap risk, use the maintenance time adjustment algorithm to reallocate the hydropower station maintenance time to avoid conflicts with the grid maintenance plan; S3.1 Obtain the time series of the hydropower station maintenance time and the grid maintenance plan; S3.2 Judge whether there is a time overlap between the hydropower station maintenance time and the grid maintenance plan; S3.3 If there is a time overlap, use the adjustment algorithm to reallocate the hydropower station maintenance time; S3.4 Determine the new maintenance time interval of the hydropower station according to the time optimization rules; S3.5 Use the time judgment logic to verify whether there is a conflict between the new maintenance time and the power grid maintenance plan; S3.6 If there is no conflict, output the reallocated maintenance time of the hydropower station; S3.7 If there is still a conflict, repeat the adjustment algorithm until the risk of time overlap is eliminated.

[0009] S4. Obtain the detailed data of the reservoir operation plan, and extract the operation requirements during the critical period and the unit power generation capacity information; S4.1 Obtain the reservoir operation plan data, and extract the reservoir volume and operation requirement values during the critical period; S4.2 Extract the unit power generation capacity value and power generation amount information, and calculate the current power generation efficiency; S4.3 If the operation requirement value is greater than the power generation capacity value, adjust the dispatching volume to meet the critical period requirements; S4.4 Calculate the difference between the planned number and the actual power generation according to the water volume value and the number of units; S4.5 Analyze the matching degree between the operation requirement value and the power generation capacity value, and judge whether dispatching adjustment is needed; S4.6 If the matching degree is lower than the preset threshold, recalculate the dispatching volume and update the planned number; S4.7 Obtain the final reservoir operation plan and power generation plan by optimizing the dispatching volume and power generation efficiency.

[0010] S5. According to the operation requirements during the critical period and the unit power generation capacity information, establish a conflict judgment model for incoming and outgoing line maintenance and reservoir dispatching, and identify the risk of unit outage caused by incoming line equipment maintenance; S5.1 Obtain the operation requirement data and unit power generation capacity data during the critical period, and establish a conflict judgment model for incoming and outgoing line maintenance and reservoir dispatching in combination with the preset maintenance plan; S5.2 According to the model analysis results, judge whether there is a conflict between equipment maintenance and reservoir dispatching. If there is a conflict, determine the risk level of unit outage caused by equipment maintenance; S5.3 Use the risk assessment algorithm to calculate the risk values of each unit for the identified outage risk, and generate a risk distribution map; S5.4 Determine the high-risk units through the risk distribution map, and judge whether there is a power generation gap in combination with the power generation capacity data; S5.5 If there is a power generation gap, use the power generation dispatching optimization algorithm to adjust the reservoir dispatching plan to reduce the unit outage risk; S5.6 According to the optimized reservoir dispatching plan, update the conflict judgment model for incoming and outgoing line maintenance and reservoir dispatching, and re-evaluate the unit outage risk; S5.7 Finally, an optimized reservoir operation plan and the assessment results of unit outage risks are obtained to provide support for decision-making.

[0011] S6. If there are unit outage risks, an in-and-out line maintenance optimization algorithm is adopted to adjust the maintenance time of the incoming line equipment to ensure that the maximum power generation capacity of the branch factory meets the reservoir operation plan; S6.1 Obtain the unit outage risk value and judge whether to trigger the maintenance optimization algorithm according to the risk value; S6.2 If triggered, extract the planned value from the reservoir operation plan, and combine it with the current power generation of the branch factory to calculate the power generation difference; S6.3 Adopt the optimization algorithm to adjust the maintenance time of the incoming line equipment to obtain the adjusted time point; S6.4 According to the adjusted time point, recalculate the power generation capacity of the branch factory and judge whether it meets the utilization rate in the reservoir operation plan; S6.5 If it meets the requirement, output the adjusted maintenance time point; S6.6 If it does not meet the requirement, re-execute the optimization algorithm until a time point that meets the utilization rate is obtained.

[0012] S7. Obtain the maintenance arrangement information of the outgoing line equipment, establish a transmission capacity judgment model, and identify the risk of insufficient transmission capacity caused by the synchronous maintenance of two outgoing line equipment; S7.1 Obtain the maintenance period and maintenance item information, and conduct data analysis on the maintenance points and maintenance lines; S7.2 Through the statistics of the maintenance quantity and frequency, determine the correlation between the maintenance sequence and the maintenance condition; S7.3 Judge the matching degree between the maintenance effect and the maintenance target, and adopt the calculation model of the maintenance scope and maintenance capacity; S7.4 If the maintenance periods overlap, identify the risk of insufficient maintenance quantity; S7.5 According to the evaluation of the maintenance condition and maintenance effect, determine the maintenance target adjustment plan; S7.6 Through the optimization of the maintenance scope and maintenance capacity, establish a transmission capacity judgment model.

[0013] S8. If there is a risk of insufficient transmission capacity, adopt the outgoing line maintenance time allocation algorithm to adjust the maintenance time of the outgoing line equipment to ensure that the transmission capacity of the normal line meets the reservoir operation plan; S8.1 Obtain the risk information of insufficient transmission capacity and judge whether the risk exists; S8.2 If there is a risk of insufficiency, adopt the outgoing line maintenance time allocation algorithm; S8.3 According to the reservoir operation plan, determine the transmission capacity requirement of the normal line; S8.4 Obtain the maintenance time data of the outgoing line equipment and adjust the maintenance time through an algorithm; S8.5 Obtain the adjusted maintenance time data of the outgoing line equipment and determine whether the transmission capacity of the normal line meets the reservoir operation plan; S8.6 If it meets the requirements, determine the final maintenance time allocation plan; S8.7 Output the maintenance time allocation plan and execute it.

[0014] S9. Obtain the maintenance history data of the busbar lines, establish a repeated maintenance judgment model, and identify the repeated maintenance situations of the busbar lines within one year; S9.1 Obtain the maintenance history data of the busbar lines, including line numbers, maintenance times, maintenance contents, and maintenance results; S9.2 Use time range screening to filter out the maintenance records within one year; S9.3 According to the line numbers and maintenance times, count the number of maintenance times for each line; S9.4 Mark the abnormal maintenance records for the maintenance contents and results; S9.5 Establish a repeated maintenance judgment model based on the number of maintenance times and abnormal marks; S9.6 If the number of maintenance times of a certain line exceeds the preset threshold and there are abnormal marks, it is determined as repeated maintenance; S9.7 Output the line numbers of the repeated maintenance, their corresponding maintenance times, and maintenance contents.

[0015] S10. If there are repeated maintenance situations, use the maintenance resource optimization algorithm to reallocate the maintenance resources of the busbar lines to avoid waste and extend the equipment life; S10.1 Obtain the historical maintenance data of the busbar lines, analyze whether there are repeated maintenance records, and if there are repeated maintenance records, extract the relevant maintenance resource allocation information; S10.2 Use the optimization algorithm model to calculate the resource reallocation plan based on the extracted maintenance resource allocation information and determine the optimal resource allocation strategy; S10.3 Generate the maintenance plan for the busbar lines according to the optimal resource allocation strategy, allocate the maintenance resources, and avoid repeated use of resources; S10.4 Obtain the equipment operation status data, combine it with the maintenance plan, and determine whether the equipment service life is affected by the maintenance. If it is affected, adjust the maintenance frequency; S10.5 Through the optimization algorithm model, calculate the impact of the adjusted maintenance frequency on resource allocation and generate a new resource allocation plan; S10.6 Update the maintenance plan for the busbar lines according to the new resource allocation plan, reduce resource waste, and extend the equipment service life; S10.7 Obtain the updated maintenance plan data, analyze the resource allocation efficiency. If there is uneven resource allocation, optimize the algorithm model again to generate the final plan.

[0016] S11. Obtain the operation data of special startup combinations, establish an operation risk judgment model, and identify the power grid operation risks under the operation modes of the left bank denominator and the right bank combined busbar; S11.1 Obtain the operation data of special startup modes, and extract relevant features for the operation modes of the left bank denominator and the right bank combined busbar; S11.2 Through a preset data processing method, clean and standardize the operation data to obtain normalized data; S11.3 Adopt a preset risk model algorithm, input the normalized data into the model, and establish a power grid operation risk judgment model; S11.4 If the data features of the left bank denominator operation mode meet the preset threshold, judge that it has a high-risk type; S11.5 If the data features of the right bank combined busbar operation mode meet the preset threshold, judge that it has a low-risk type; S11.6 According to the risk judgment results, generate the risk level classification of the left bank denominator and the right bank combined busbar operation modes; S11.7 Through the risk level classification output by the model, determine the priority of the power grid operation risks and complete the risk identification task.

[0017] S12. If there are operation risks, adopt a startup combination optimization algorithm to adjust the operation mode to ensure that the maximum power generation capacity of the power station meets the reservoir operation plan; S12.1 Obtain the reservoir operation plan and the power station operation data, and identify potential operation risks; S12.2 Adopt a risk identification algorithm to judge whether the current operation mode meets the requirements of the reservoir plan; S12.3 If there are risks, generate a startup combination adjustment plan through an optimization algorithm; S12.4 According to the adjustment plan, reconfigure the power station operation parameters to optimize the power generation capacity; S12.5 Adopt a capacity matching algorithm to verify whether the power generation capacity of the power station meets the reservoir plan requirements; S12.6 If the matching is successful, execute the adjusted operation mode to control the operation risks; S12.7 Real-time monitor the operation status to ensure the continuous effectiveness of the adjustment plan and the controllability of risks.

[0018] The present invention can achieve the following beneficial effects: 1. By constructing a variety of targeted judgment models, the present invention can accurately identify complex and diverse potential risks such as overlapping maintenance times, unit outages, insufficient transmission capacity, repeated maintenance, and power grid operation, providing a reliable basis for subsequent decision-making and avoiding serious accidents and losses caused by risk neglect.

[0019] 2. By using various optimization algorithms, scientific adjustments are made to maintenance times, operation modes, and resource allocation, effectively reducing maintenance conflicts, avoiding the risk of overlapping times, ensuring a compact and reasonable maintenance plan, and improving maintenance efficiency and resource utilization.

[0020] 3. Guided by the reservoir operation plan, by optimizing the maintenance times of incoming and outgoing lines and adjusting the operation mode, it effectively guarantees that the maximum power generation capacity of the power station meets the reservoir operation requirements, stabilizes power supply, improves the overall operation efficiency of the hydropower station, and promotes the efficient utilization of hydropower resources.

[0021] 4. Regarding the problem of repeated maintenance of busbar lines, optimize the allocation of maintenance resources, avoid excessive maintenance, reduce equipment wear, effectively extend the service life of equipment, reduce equipment maintenance costs and replacement frequencies, and enhance the reliability and stability of the operation of hydropower station equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention will be further described below in conjunction with the drawings and embodiments: Figure 1 is the method logic diagram of the present invention; Figure 2 is the flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The preferred solution is as Figures 1 to 2 shown, a method for risk identification and reminder of the annual maintenance plan of a hydropower station, including: S1. Obtain detailed data on the maintenance requirements of hydropower station equipment and the grid maintenance plan, and extract time node information.

[0024] Obtain the maintenance data of hydropower station equipment and the grid maintenance plan, extract the maintenance time nodes and equipment status information. According to the maintenance time nodes and equipment status, judge the maintenance priority and maintenance type. Use the preset maintenance cycle and grid load data to determine the maintenance duration and maintenance location. Through the maintenance priority and maintenance type, match the maintenance personnel and maintenance tool information. If the maintenance duration exceeds the preset threshold, adjust the grid load data and the maintenance plan. According to the adjusted maintenance plan, update the maintenance time nodes and maintenance priorities. Obtain the updated maintenance plan data and generate the final maintenance arrangement information.

[0025] Specifically, by analyzing the detailed data of the maintenance requirements of hydropower station equipment and the grid maintenance plan, time node information is extracted. First, it is necessary to obtain the maintenance cycle and the last maintenance time of the equipment from the maintenance requirements of hydropower station equipment. For example, the maintenance cycle of the water turbine of a certain hydropower station is 12 months, and the last maintenance time is January 2023. Then the next maintenance time is expected to be January 2024. At the same time, obtain the grid maintenance time from the grid maintenance plan. For example, a certain grid plans to conduct a 10-day maintenance in December 2023. By comparing the maintenance time of hydropower station equipment with the grid maintenance time, it is found that there is an overlap between the maintenance time of hydropower station equipment and the grid maintenance time. Therefore, it is necessary to adjust the maintenance time of hydropower station equipment to avoid conflicts. Through algorithm calculation, the maintenance time of hydropower station equipment is adjusted to February 2024 to ensure that the maintenance of hydropower station equipment does not conflict with the grid maintenance. Further analysis shows that if the maintenance time of hydropower station equipment cannot be adjusted, it is necessary to coordinate the grid maintenance time. For example, advance the grid maintenance time to November 2023. In this way, ensure the smooth progress of the maintenance of hydropower station equipment and the grid maintenance, and at the same time ensure the stable operation of the grid.

[0026] S2. According to the time node information, establish an overlap judgment model for maintenance time to identify the risk of time overlap between the maintenance of the hydropower station bus and the maintenance of the grid converter station bus.

[0027] Obtain the maintenance time points of the hydropower station bus and the maintenance time points of the grid converter station bus, generate two independent time lines, and mark the start and end times of the maintenance period respectively. Judge whether there is an intersection between the two time lines through a preset model. If there is an intersection, it is determined that the maintenance periods overlap. For the overlapping maintenance periods, extract the specific time periods on the time axis and mark them as risk points. According to the time range of the risk points, generate an overlap risk report, including the maintenance item information of the hydropower station and the converter station. Use machine learning algorithms to analyze historical maintenance data and establish a time point prediction model for optimizing the maintenance time arrangement. Through the time point prediction model, adjust the maintenance time points of the hydropower station and the converter station to reduce the overlap risk. According to the adjusted time points, regenerate the time line and conduct an overlap risk judgment again to obtain the final optimization result.

[0028] S3. If there is a risk of time overlap, use a maintenance time adjustment algorithm to reallocate the maintenance time of the hydropower station to avoid conflicts with the grid maintenance plan.

[0029] Obtain the time series of the maintenance time of the hydropower station and the grid maintenance plan. Determine whether there is a time overlap between the maintenance time of the hydropower station and the grid maintenance plan. If there is a time overlap, use an adjustment algorithm to reallocate the maintenance time of the hydropower station. According to the time optimization rules, determine the new maintenance time interval of the hydropower station. Use the time judgment logic to verify whether there is a conflict between the new maintenance time and the grid maintenance plan. If there is no conflict, output the reallocated maintenance time of the hydropower station. If there is still a conflict, repeat the adjustment algorithm until the risk of time overlap is eliminated.

[0030] Specifically, in the case of a conflict between the hydropower station maintenance and the grid maintenance plan, it is first necessary to collect the maintenance time data of the hydropower station and the grid. For example, Hydropower Station A plans to conduct maintenance from October 1st to October 15th, 2023, while Grid B plans to conduct maintenance from October 10th to October 20th, 2023. Through the time overlap detection algorithm, it can be identified that there is an overlap between the two from October 10th to October 15th, 2023. To avoid conflicts, an adjustment algorithm for maintenance time is used. For example, the maintenance time of Hydropower Station A is advanced to September 20th to October 5th, 2023. This adjustment is based on the priority of the hydropower station and the non-adjustability of the grid. Through the priority sorting algorithm, it is determined that the priority of Hydropower Station A is high and the priority of Grid B is low. During the adjustment process, it is also necessary to consider the power generation capacity of the hydropower station and the load demand of the grid. For example, the power generation capacity of Hydropower Station A from September 20th to October 5th is 500 MW, while the load demand of Grid B from October 10th to October 20th is 600 MW. Through the load balancing algorithm, it can be ensured that the early maintenance of Hydropower Station A will not affect the load of the grid. Finally, the adjusted maintenance plan is automatically updated by the system and relevant management personnel are notified. The entire process is automatically processed by information technology to ensure that the maintenance plans of the hydropower station and the grid are conflict-free and efficiently executed.

[0031] S4. Obtain the detailed data of the reservoir operation plan, and extract the operation requirements during the critical period and the unit power generation capacity information.

[0032] Obtain the reservoir operation plan data, and extract the reservoir volume and operation demand values during the critical period. Extract the unit power generation capacity value and the power generation amount information, and calculate the current power generation efficiency. If the operation demand value is greater than the power generation capacity value, adjust the dispatching volume to meet the critical period demand. Calculate the difference between the planned number and the actual power generation amount based on the water volume value and the number of units. Analyze the matching degree between the operation demand value and the power generation capacity value, and judge whether dispatching adjustment is needed. If the matching degree is lower than the preset threshold, recalculate the dispatching volume and update the planned number. Through optimizing the dispatching volume and the power generation efficiency, obtain the final reservoir operation plan and power generation plan.

[0033] Specifically, when obtaining the detailed data of the reservoir operation plan, basic data such as historical water levels, flows, and water storage volumes were first extracted through the reservoir management system, and combined with meteorological forecast information to predict the inflow for the next month. For example, according to historical data analysis, the average inflow of the reservoir during the flood season is 500 cubic meters per second, while it drops to 200 cubic meters per second during the dry season. Through time series analysis, the ARIMA model was used to predict the inflow for the next 30 days. The prediction results show that the inflow for the next week is 450 cubic meters per second, and then gradually decreases to 300 cubic meters per second. Next, the operation requirement information during the critical period was extracted. The operation requirements of the reservoir during the flood control period mainly focus on controlling the water level not to exceed the warning line, and the warning water level is 500 meters above sea level. While during the water supply period, the operation requirement is to ensure that the lowest water level is not lower than 450 meters above sea level to meet the downstream agricultural irrigation and urban water supply needs. Regarding the information on the generating capacity of the units, according to the water level and flow data of the reservoir, the generating power of the water turbines was calculated. For example, when the reservoir water level is 480 meters above sea level and the flow is 400 cubic meters per second, the generating power of the water turbine is 200 megawatts, and when the water level drops to 460 meters above sea level and the flow is 300 cubic meters per second, the generating power drops to 150 megawatts. Through the above analysis, the generating capacity range of the units under different operation requirements can be determined, and combined with the inflow prediction, the reservoir operation plan can be optimized to ensure flood control and water supply requirements while maximizing the power generation benefit.

[0034] S5. According to the operation requirements during the critical period and the information on the generating capacity of the units, establish a conflict judgment model for the maintenance of incoming and outgoing lines and reservoir operation, and identify the risk of unit outage caused by the maintenance of incoming line equipment.

[0035] Obtain the operation requirement data during the critical period and the generating capacity data of the units, and combine with the preset maintenance plan to establish a conflict judgment model for the maintenance of incoming and outgoing lines and reservoir operation. According to the analysis results of the model, judge whether there is a conflict between equipment maintenance and reservoir operation. If there is a conflict, determine the risk level of unit outage caused by equipment maintenance. Use the risk assessment algorithm to calculate the risk values of each unit for the identified outage risk and generate a risk distribution map. Through the risk distribution map, determine the high-risk units, and combine with the generating capacity data to judge whether there is a power generation gap. If there is a power generation gap, use the power generation scheduling optimization algorithm to adjust the reservoir operation plan to reduce the risk of unit outage. According to the optimized reservoir operation plan, update the conflict judgment model for the maintenance of incoming and outgoing lines and reservoir operation, and re-evaluate the risk of unit outage. Finally, obtain the optimized reservoir operation plan and the risk assessment result of unit outage to provide support for decision-making.

[0036] Specifically, to establish a conflict judgment model for the maintenance of incoming and outgoing lines and reservoir operation, it is first necessary to obtain the operation requirements during the critical period and the generating capacity information of the units. For example, the operation requirement of a hydropower station during the critical period is that the daily power generation is not less than 5 million kWh, and the maximum generating capacity of the unit is 300,000 kWh per hour. Based on these data, the minimum operation time of the unit during the critical period can be calculated, that is, 5 million kWh / 300,000 kWh / hour ≈ 167 hours. Next, combined with the reservoir operation plan, analyze the impact of the change in reservoir water level on the generating capacity. Assume that the water level change range of the reservoir during the critical period is from 100 meters to 120 meters. Through the turbine efficiency curve, the generating efficiency of the unit at different water levels can be obtained. For example, when the water level is 100 meters, the unit efficiency is 90%, and the generating capacity is 270,000 kWh / hour; when the water level is 120 meters, the unit efficiency is 95%, and the generating capacity is 250,000 kWh / hour. Then, according to the maintenance plan of the incoming line equipment, judge whether the unit will be shut down during the maintenance period. Assume that the maintenance time of a certain incoming line equipment is 8 hours, and the unit cannot operate during the maintenance period, then the power generation loss during the maintenance period is 8 hours × 270,000 kWh / hour = 2.16 million kWh. By comparing the power generation loss with the operation requirements during the critical period, the risk of unit shutdown caused by maintenance can be identified. If the power generation loss exceeds 10% of the operation requirements during the critical period (i.e., 500,000 kWh), it is considered a high risk. In this example, 2.16 million kWh far exceeds 500,000 kWh, indicating that this maintenance plan has a high risk and it is necessary to adjust the maintenance time or take other measures to reduce the power generation loss.

[0037] S6. If there is a risk of unit shutdown, adopt the incoming and outgoing line maintenance optimization algorithm to adjust the maintenance time of the incoming line equipment to ensure that the maximum generating capacity of the branch factory meets the reservoir operation plan.

[0038] Obtain the unit shutdown risk value, and judge whether to trigger the maintenance optimization algorithm according to the risk value. If triggered, extract the planned value from the reservoir operation plan, combine it with the current power generation of the branch factory, and calculate the power generation difference. Adopt the optimization algorithm to adjust the maintenance time of the incoming line equipment to obtain the adjusted time point. According to the adjusted time point, recalculate the generating capacity of the branch factory and judge whether it meets the utilization rate in the reservoir operation plan. If it meets, output the adjusted maintenance time point; if it does not meet, re-execute the optimization algorithm until a time point that meets the utilization rate is obtained.

[0039] Specifically, in the power system, if there is a risk of unit outage, the maintenance time of the incoming line equipment can be adjusted through the incoming and outgoing line maintenance optimization algorithm to ensure that the maximum power generation capacity of the branch factory meets the reservoir operation plan. First, the system will monitor the operating status of each unit in real time and predict possible outage risks in the future. For example, a certain unit has a 20% outage probability within the next 30 days due to equipment aging. Based on this prediction, the system will start the maintenance optimization algorithm, which comprehensively considers the water level changes in the reservoir, power generation demand, and the urgency of equipment maintenance. Suppose the water level in the reservoir will drop by 10% within the next 15 days, and the current power generation demand of the branch factory is 500 megawatts. The system will calculate the optimal maintenance time window based on these data. Through the linear programming model, the system determines to schedule the maintenance time 5 days later and last for 3 days, which can minimize the impact on power generation capacity. Specifically, the system will calculate that during these 3 days of maintenance, the power generation capacity of the branch factory will be reduced to 400 megawatts, but it can still meet the minimum power generation demand of 450 megawatts of the reservoir operation plan. In addition, the system will also consider the spare capacity of other units to ensure that there are enough spare units to make up for the power generation gap during the maintenance period. For example, another unit in the system can provide 50 megawatts of spare capacity, further reducing the risk of insufficient power generation capacity. Through this optimization algorithm, the system can maximize the power generation capacity of the branch factory while ensuring the safe operation of the equipment and meeting the requirements of the reservoir operation plan.

[0040] S7. Obtain the maintenance arrangement information of the outgoing line equipment, establish a transmission capacity judgment model, and identify the risk of insufficient transmission capacity caused by the synchronous maintenance of two outgoing line equipments.

[0041] Obtain the maintenance period and maintenance item information, and conduct data analysis on the maintenance points and maintenance lines. Through the statistics of the maintenance volume and maintenance frequency, determine the correlation between the maintenance sequence and the maintenance situation. Judge the matching degree between the maintenance effect and the maintenance target, and adopt the calculation model of the maintenance scope and maintenance power. If the maintenance periods overlap, identify the risk of insufficient maintenance volume. Based on the evaluation of the maintenance situation and maintenance effect, determine the maintenance target adjustment plan. Through the optimization of the maintenance scope and maintenance power, establish a transmission capacity judgment model.

[0042] Specifically, when obtaining the maintenance schedule information of outgoing line equipment, first extract the maintenance plans of each outgoing line equipment from the power system, including the start time, end time, and maintenance type, etc. For example, the maintenance plans of two outgoing line equipment A and B in a certain substation are as follows: Equipment A conducts annual maintenance from October 1st to October 5th, 2023, and equipment B conducts fault maintenance from October 3rd to October 7th, 2023. Through time series analysis, it is identified that the maintenance times of these two pieces of equipment overlap, that is, from October 3rd to October 5th. In order to establish a transmission capacity judgment model, a regression analysis method based on historical data is used to calculate the maximum transmission capacity of the outgoing line equipment under normal conditions. For example, the maximum transmission capacity of equipment A is 500 MW, and the maximum transmission capacity of equipment B is 600 MW. During the maintenance period, the transmission capacity of equipment A drops to 0, and the transmission capacity of equipment B drops to 300 MW. Through superposition analysis, it is found that during the synchronous maintenance of the two pieces of equipment, the total transmission capacity of the system drops from 1100 MW to 300 MW, far lower than the minimum transmission capacity of 800 MW required for the normal operation of the system. In order to further identify risks, the Monte Carlo simulation method is used to simulate the changes in the system transmission capacity under different maintenance plans. For example, if the maintenance time of equipment A is adjusted to October 6th to October 10th, the total transmission capacity of the system from October 3rd to October 5th will remain at 900 MW, meeting the system requirements. Through the above analysis, the risk of insufficient transmission capacity caused by the synchronous maintenance of two outgoing line equipment can be effectively identified and avoided, ensuring the stable operation of the power system.

[0043] S8. If there is a risk of insufficient transmission capacity, adopt the outgoing line maintenance time allocation algorithm to adjust the maintenance time of the outgoing line equipment to ensure that the transmission capacity of the normal line meets the reservoir operation plan.

[0044] Obtain the risk information of insufficient transmission capacity and judge whether the risk exists. If there is a risk of insufficiency, then adopt the outgoing line maintenance time allocation algorithm. According to the reservoir operation plan, determine the transmission capacity requirement of the normal line. Obtain the maintenance time data of the outgoing line equipment and adjust the maintenance time through the algorithm. Obtain the adjusted maintenance time data of the outgoing line equipment and judge whether the transmission capacity of the normal line meets the reservoir operation plan. If it meets, determine the final maintenance time allocation plan. Output the maintenance time allocation plan and execute it.

[0045] Specifically, in actual operation, first, the operation status data of the outgoing line equipment are collected, including parameters such as current, voltage, and power, and the future transmission capacity is evaluated using a load forecasting algorithm. For example, the time series analysis method is adopted, combined with historical data and weather factors, to predict the load demand for the next week. If the prediction result shows that the transmission capacity of a certain line may be insufficient, the system will automatically start the outgoing line maintenance time allocation algorithm. This algorithm is based on a genetic optimization model, considering the equipment maintenance duration, maintenance priority, and line load conditions, to generate an optimal maintenance schedule. For example, if the maintenance duration of a certain line is 8 hours, the system will arrange it during the early morning period with lower load to ensure that the transmission capacity of other lines can meet the reservoir operation plan during the maintenance period. Through real-time monitoring and adjustment, the system can dynamically optimize the maintenance plan to ensure the stable operation of the power grid. For example, if the load of a certain line suddenly increases, the system will recalculate and adjust the maintenance time to avoid affecting the normal operation of the reservoir. The entire process ensures the stability and reliability of the transmission capacity through automated data processing and algorithm optimization.

[0046] S9. Obtain the maintenance historical data of the bus line, establish a repeated maintenance judgment model, and identify the repeated maintenance situation of the bus line within one year.

[0047] Obtain the maintenance historical data of the bus line, including the line number, maintenance time, maintenance content, and maintenance result. Use the time range screening to filter out the maintenance records within one year. According to the line number and maintenance time, count the maintenance times of each line. For the maintenance content and result, mark the abnormal maintenance records. Establish a repeated maintenance judgment model based on the maintenance times and abnormal marks. If the maintenance times of a certain line exceed the preset threshold and there are abnormal marks, it is determined as repeated maintenance. Output the line numbers of the repeated maintenance, their corresponding maintenance times, and maintenance content.

[0048] Specifically, first, obtain the maintenance historical data of the bus line from the power system, including fields such as maintenance date, maintenance reason, and maintenance duration. For example, the maintenance records of a certain bus line in a year may include maintenance for 2 hours due to overload on January 15, 2023, maintenance for 3 hours due to equipment aging on March 20, 2023, and maintenance for 1 hour due to short circuit on July 10, 2023. Next, use the K-means clustering algorithm to perform clustering analysis on these maintenance data to identify repeated maintenance patterns. Set the number of clusters to 3, and divide the maintenance records into three categories by calculating the Euclidean distance between each maintenance record. By analyzing the clustering results, it is found that the maintenance records on January 15, 2023, and March 20, 2023, are classified into the same category, indicating that the reasons for these two maintenance operations are similar and there may be repeated maintenance. For further verification, use the support vector machine (SVM) algorithm to establish a repeated maintenance judgment model, take the maintenance reason, maintenance duration, etc. as features, and take whether it is repeated maintenance as the label. Through training the model, a judgment result with an accuracy of 85% is obtained. Finally, use this model to predict the annual maintenance data and identify that there are 5 repeated maintenance operations in 2023, mainly concentrated on overload and equipment aging problems. Through these steps, the repeated maintenance situation of the bus line within a year can be effectively identified, providing data support for optimizing the maintenance strategy.

[0049] S10. If there is a repeated maintenance situation, adopt a maintenance resource optimization algorithm to reallocate the maintenance resources of the bus line, avoid waste, and extend the equipment life.

[0050] Obtain the historical maintenance data of the bus line, analyze whether there are repeated maintenance records, and if there are repeated maintenance records, extract the relevant maintenance resource allocation information. Adopt an optimization algorithm model, calculate the resource reallocation plan according to the extracted maintenance resource allocation information, and determine the optimal resource allocation strategy. According to the optimal resource allocation strategy, generate a maintenance plan for the bus line, allocate maintenance resources, and avoid repeated use of resources. Obtain the equipment operation status data, combine it with the maintenance plan, and judge whether the equipment service life is affected by the maintenance. If it is affected, adjust the maintenance frequency. Through the optimization algorithm model, calculate the impact of the adjusted maintenance frequency on resource allocation, and generate a new resource allocation plan. According to the new resource allocation plan, update the maintenance plan of the bus line, reduce resource waste, and extend the equipment service life. Obtain the updated maintenance plan data, analyze the resource allocation efficiency, and if there is uneven resource allocation, optimize the algorithm model again to generate the final plan.

[0051] Specifically, in the case of repeated maintenance, the maintenance resource optimization algorithm can be used to reallocate the maintenance resources of the bus lines, avoiding waste and extending the equipment life. First, based on historical maintenance data, the K-means clustering algorithm is used to classify the bus lines, grouping the lines with similar maintenance frequencies and fault characteristics into the same category. For example, the lines with a maintenance frequency higher than 3 times a year are classified into the high-maintenance-demand group, and the lines with a maintenance frequency lower than 1 time a year are classified into the low-maintenance-demand group. Then, the genetic algorithm is used to optimize the allocation of maintenance resources for the high-maintenance-demand group, setting the objective function as minimizing the maintenance cost and maximizing the equipment life. The maintenance cost includes labor cost and equipment loss cost, and the equipment life is quantified through the relationship between the equipment operation time and the failure rate. Through iterative optimization, the genetic algorithm outputs the optimal maintenance resource allocation plan. For example, the lines that were originally maintained once a quarter are adjusted to be maintained once every six months, while increasing the input of maintenance resources to ensure the maintenance quality. For the low-maintenance-demand group, a prediction model based on time series analysis is used to predict the future operation state of the equipment. For example, the ARIMA model is used to predict the failure probability of the equipment in the next 6 months. If the predicted value is lower than 0.5, the maintenance cycle is appropriately extended, from once a year to once every two years. In addition, combined with the equipment operation environment data, such as temperature, humidity, etc., the multiple linear regression model is used to further optimize the equipment life. For example, in a high-temperature and high-humidity environment, the maintenance cycle is shortened by 10% to reduce the failure risk. Finally, the optimized maintenance plan is combined with the real-time monitoring system. Through the Internet of Things technology, the equipment operation data is collected in real time, and the support vector machine (SVM) algorithm is used to classify the equipment state. When the equipment state reaches the warning threshold, the maintenance task is automatically triggered to ensure the scientificity and timeliness of resource allocation.

[0052] S11. Obtain the operation data of the special startup combination, establish an operation risk judgment model, and identify the power grid operation risks under the operation modes of the left-bank denominator and the right-bank combined bus.

[0053] Obtain the operation data of the special startup mode. For the operation modes of the left-bank denominator and the right-bank combined bus, extract the relevant features. Through the preset data processing method, the operation data is cleaned and standardized to obtain the normalized data. Using the preset risk model algorithm, the normalized data is input into the model to establish a power grid operation risk judgment model. If the data features of the left-bank denominator operation mode meet the preset threshold, it is judged that there is a high-risk type. If the data features of the right-bank combined bus operation mode meet the preset threshold, it is judged that there is a low-risk type. According to the risk judgment results, generate the risk level classification of the operation modes of the left-bank denominator and the right-bank combined bus. Through the risk level classification output by the model, determine the priority of the power grid operation risks and complete the risk identification task.

[0054] Specifically, when obtaining the operation data of the special startup combination, first collect the real-time data of the left bank denominator and the right bank combined bus operation modes through the power grid monitoring system, including parameters such as voltage, current, and power factor. For example, in the left bank denominator operation mode, the voltage is 220 kV, the current is 1500 A, and the power factor is 95; in the right bank combined bus operation mode, the voltage is 230 kV, the current is 1400 A, and the power factor is 93. These data are cleaned and normalized through the data preprocessing module to ensure the accuracy and consistency of the data. Next, use machine learning algorithms to establish an operation risk judgment model, select the support vector machine (SVM) as the classifier, use the radial basis function (RBF) as the kernel function, set the penalty parameter C to 0, and set the kernel function parameter γ to 1. During the model training process, use historical data as the training set and optimize the model parameters through the cross-validation method to ensure the generalization ability of the model. In the model evaluation stage, use accuracy, recall rate, and F1 value as evaluation indicators. For example, the accuracy of the model reaches 95%, the recall rate is 92%, and the F1 value is 93%. Finally, based on the established model, identify the power grid operation risks in the left bank denominator and the right bank combined bus operation modes. The risk score in the left bank denominator operation mode is 85, and the risk score in the right bank combined bus operation mode is 78. Through the risk score, the power grid operation risks under different operation modes can be intuitively judged, providing a scientific basis for power grid dispatching.

[0055] S12. If there is an operation risk, adopt the startup combination optimization algorithm to adjust the operation mode to ensure that the maximum power generation capacity of the power station meets the reservoir operation plan.

[0056] Obtain the reservoir operation plan and the power station operation data, and identify potential operation risks. Adopt the risk identification algorithm to judge whether the current operation mode meets the requirements of the reservoir plan. If there is a risk, generate a startup combination adjustment plan through the optimization algorithm. According to the adjustment plan, reconfigure the power station operation parameters to optimize the power generation capacity. Adopt the capacity matching algorithm to verify whether the power station power generation capacity meets the reservoir plan requirements. If the matching is successful, execute the adjusted operation mode to control the operation risk. Monitor the operation status in real time to ensure the continuous effectiveness of the adjustment plan and the controllability of the risk.

[0057] Specifically, in the presence of operation risks, the start-up unit combination optimization algorithm can effectively adjust the operation mode to ensure that the maximum power generation capacity of the power station meets the reservoir operation plan. First, through historical data analysis, the variation law of the power generation efficiency of the power station under different water levels and flow conditions is identified. For example, when the reservoir water level is between 150 meters and 160 meters, the power generation efficiency can reach up to 95%, while when the water level is below 140 meters, the efficiency drops to 85%. Based on these data, a multi-objective optimization model is constructed, and the objective function includes maximizing power generation, minimizing operation costs, and risks. The genetic algorithm is used for solution, with the population size set to 100, the number of iterations set to 500, the crossover probability set to 8, and the mutation probability set to 1. During the optimization process, the power generation capacity and risk level of different start-up unit combinations are evaluated in each iteration, and the optimal solution is selected. For example, in one iteration, it is found that when two units operate simultaneously, the power generation can reach 200 megawatts, and the risk level is controlled below 2%. Further analysis shows that this combination not only meets the reservoir operation plan but also effectively reduces operation risks. Through continuous optimization and adjustment, it is finally determined that under specific water level and flow conditions, the optimal start-up unit combination is for three units to operate simultaneously, with the power generation reaching 300 megawatts and the risk level remaining below 5%. This result ensures a high degree of matching between the maximum power generation capacity of the power station and the reservoir operation plan while minimizing operation risks to the greatest extent.

[0058] The above embodiments are only the preferred technical solutions of the present invention and should not be regarded as limitations on the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for identifying and prompting risks in the annual maintenance plan of a hydropower station, characterized in that The following steps are involved: Obtain detailed data on hydropower station equipment maintenance requirements and power grid maintenance plans, and extract time node information; Based on the time node information, a maintenance time overlap judgment model is established to identify the time overlap risk between the busbar maintenance of the hydropower station and the busbar maintenance of the power grid converter station; If there is a risk of time overlap, the maintenance time adjustment algorithm is used to reallocate the maintenance time of the hydropower station to avoid conflicts with the maintenance plan of the power grid; Obtain detailed data on reservoir operation plans and extract information on critical period operation requirements and unit power generation capacity; According to the operation demand during the critical period and the generating capacity information of the unit, a conflict judgment model between incoming and outgoing line maintenance and reservoir scheduling is established to identify the risk of unit shutdown caused by incoming line equipment maintenance; If there is a risk of unit shutdown, the incoming and outgoing line maintenance optimization algorithm is used to adjust the incoming line equipment maintenance time to ensure that the maximum power generation capacity of the branch plant meets the reservoir operation plan; Obtain outgoing line equipment maintenance schedule information, establish an outgoing line capacity judgment model, and identify the risk of insufficient outgoing line capacity caused by simultaneous maintenance of two outgoing line equipment; If there is a risk of insufficient transmission capacity, the outgoing line maintenance time allocation algorithm is used to adjust the outgoing line equipment maintenance time to ensure that the transmission capacity of the normal line meets the reservoir operation plan; Obtain the historical maintenance data of the bus line, establish a repeated maintenance judgment model, and identify the repeated maintenance of the bus line within a year; If repeated maintenance occurs, a maintenance resource optimization algorithm is used to reallocate bus line maintenance resources to avoid waste and extend equipment life; Obtain the operation data of special startup combinations, establish an operation risk judgment model, and identify the operation risks of the power grid under the left bank denominator and right bank combined denominator operation modes; If there are operational risks, the startup combination optimization algorithm is used to adjust the operation mode to ensure that the maximum power generation capacity of the plant meets the reservoir utilization plan.

2. The method according to claim 1, characterized in that: The acquisition of detailed data on the maintenance requirements of hydropower station equipment and the power grid maintenance plan, and extraction of time node information, includes: Obtain hydropower station equipment maintenance data and power grid maintenance plan, and extract maintenance time nodes and equipment status information; Determine maintenance priority and maintenance type based on maintenance time nodes and equipment status; Use preset maintenance cycles and grid load data to determine maintenance duration and location; Match maintenance personnel and maintenance tool information based on maintenance priority and maintenance type; If the maintenance duration exceeds the preset threshold, the grid load data and maintenance plan are adjusted; Update the maintenance time nodes and maintenance priorities according to the adjusted maintenance plan; Obtain updated maintenance plan data and generate final maintenance schedule information.

3. The method according to claim 1, characterized in that The maintenance time overlap judgment model is established based on the time node information to identify the time overlap risk of the busbar maintenance of the hydropower station and the busbar maintenance of the power grid converter station, including: Obtain the maintenance time points of the busbars of the hydropower station and the power grid converter station, generate two independent timelines, and mark the start and end time of the maintenance period respectively; A preset model is used to determine whether the two timelines have an intersection. If so, it is determined that the maintenance periods overlap; For overlapping maintenance periods, extract specific time periods on the timeline and mark them as risk points; Generate overlapping risk reports based on the time range of risk points, including maintenance item information for hydropower stations and converter stations; Use machine learning algorithms to analyze historical maintenance data and establish a time point prediction model to optimize maintenance schedules; Through the time point prediction model, the maintenance time points of hydropower stations and converter stations are adjusted to reduce the risk of overlap; Based on the adjusted time points, the timeline is regenerated, and the overlapping risk judgment is performed again to obtain the final optimization result.

4. The method according to claim 1, characterized in that: If there is a risk of time overlap, the maintenance time adjustment algorithm is used to reallocate the maintenance time of the hydropower station to avoid conflicts with the maintenance plan of the power grid, including: Obtain the time series of hydropower station maintenance time and power grid maintenance plan; Determine whether there is any overlap between the maintenance schedule of the hydropower station and the maintenance schedule of the power grid; If there is time overlap, an adjustment algorithm is used to reallocate the maintenance time of the hydropower station; According to the time optimization rules, determine the new maintenance time interval of the hydropower station; Use time judgment logic to verify whether the new maintenance time conflicts with the power grid maintenance plan; If there is no conflict, then the reallocated maintenance time of the hydropower station is output; If there is still a conflict, the algorithm is adjusted repeatedly until the risk of time overlap is eliminated.

5. The method according to claim 1, characterized in that The detailed data of the reservoir operation plan is obtained, and the operation demand and power generation capacity information of the unit during the critical period are extracted, including: Obtain reservoir operation plan data and extract reservoir volume and operation demand values ​​during key periods; Extract the generating capacity and power generation information of the unit and calculate the current power generation efficiency; If the operation demand value is greater than the power generation capacity value, the dispatch quantity is adjusted to meet the demand during the critical period; Calculate the difference between the planned and actual power generation based on the water volume value and the number of units; Analyze the matching degree between the operation demand value and the power generation capacity value to determine whether scheduling adjustment is needed; If the matching degree is lower than the preset threshold, the scheduling quantity is recalculated and the planned number is updated; By optimizing the dispatching volume and power generation efficiency, the final reservoir operation plan and power generation plan are obtained.

6. The method according to claim 1, characterized in that According to the operation demand in the critical period and the generating capacity information of the unit, a conflict judgment model between the inlet and outlet line maintenance and the reservoir scheduling is established to identify the unit shutdown risk caused by the inlet line equipment maintenance, including: Obtain the operation demand data and unit power generation capacity data during the critical period, and establish a conflict judgment model between inlet and outlet line maintenance and reservoir scheduling in combination with the preset maintenance plan; Based on the model analysis results, determine whether there is a conflict between equipment maintenance and reservoir scheduling. If there is a conflict, determine the risk level of unit shutdown caused by equipment maintenance; Using risk assessment algorithms, we calculate the risk value of each unit based on the identified outage risks and generate a risk distribution map; Through the risk distribution map, we can identify high-risk units and combine them with the power generation capacity data to determine whether there is a power generation gap; If there is a power generation gap, the power generation scheduling optimization algorithm is used to adjust the reservoir scheduling plan to reduce the risk of unit shutdown; According to the optimized reservoir dispatch plan, the conflict judgment model between inlet and outlet line maintenance and reservoir dispatch is updated, and the risk of unit shutdown is re-evaluated; Finally, the optimized reservoir scheduling plan and unit outage risk assessment results are obtained to provide support for decision-making.

7. The method according to claim 1, characterized in that If there is a risk of unit shutdown, the inlet and outlet maintenance optimization algorithm is used to adjust the maintenance time of the inlet equipment to ensure that the maximum power generation capacity of the branch plant meets the reservoir operation plan, including: Obtain the unit outage risk value and determine whether to trigger the maintenance optimization algorithm based on the risk value; If triggered, the planned value is extracted from the reservoir operation plan, combined with the current power generation of the branch plant, to calculate the power generation difference; Adopt optimization algorithm to adjust the maintenance time of incoming equipment and obtain the adjusted time point; Recalculate the power generation capacity of the branch plant according to the adjusted time point to determine whether the utilization rate in the reservoir utilization plan is met; If satisfied, output the adjusted maintenance time point; If not, re-execute the optimization algorithm until a time point that satisfies the utilization rate is obtained.

8. The method according to claim 1, characterized in that The obtaining of outgoing line equipment maintenance arrangement information, establishing an outgoing line capacity judgment model, and identifying the risk of insufficient outgoing line capacity caused by simultaneous maintenance of two outgoing line equipments include: Obtain maintenance period and maintenance item information, and perform data analysis on maintenance points and maintenance lines; Through the statistics of maintenance quantity and maintenance frequency, the correlation between maintenance sequence and maintenance status is determined; To judge the matching degree between maintenance effect and maintenance target, a calculation model of maintenance standard and maintenance force is adopted; If the maintenance periods overlap, identify the risk of insufficient maintenance; Determine the maintenance standard adjustment plan based on the evaluation of maintenance status and maintenance effectiveness; By optimizing the maintenance range and maintenance force, a delivery capacity judgment model is established.

9. The method according to claim 1, characterized in that: If there is a risk of insufficient delivery capacity, the outgoing line maintenance time allocation algorithm is used to adjust the outgoing line equipment maintenance time to ensure that the delivery capacity of the normal line meets the reservoir operation plan, including: Obtain information on the risk of insufficient delivery capacity and determine whether the risk exists; If there is a risk of insufficiency, the outgoing line maintenance time allocation algorithm is used; Determine the delivery capacity requirements of normal lines according to the reservoir operation plan; Obtain maintenance time data of outgoing equipment and adjust maintenance time through algorithms; Obtain the adjusted outgoing line equipment maintenance time data to determine whether the normal line delivery capacity meets the reservoir operation plan; If satisfied, the final maintenance time allocation plan is determined; Output the maintenance time allocation plan and execute it.

10. The method according to claim 1, characterized in that The acquisition of the bus line maintenance history data, establishment of a repeated maintenance judgment model, and identification of repeated maintenance situations of the bus line within one year include: Obtain the maintenance history data of the bus line, including line number, maintenance time, maintenance content and maintenance results; Use time range screening to filter out maintenance records within one year; According to the line number and maintenance time, count the number of maintenance times of each line; Mark abnormal maintenance records according to maintenance content and maintenance results; Establish a repeated maintenance judgment model based on maintenance times and abnormality marks; If the number of inspections of a line exceeds the preset threshold and there is an abnormal mark, it is judged as repeated inspection; Output the line number of repeated maintenance and its corresponding maintenance time and maintenance content.

11. The method according to claim 1, characterized in that: If repeated maintenance occurs, a maintenance resource optimization algorithm is used to reallocate bus line maintenance resources to avoid waste and extend equipment life, including: Obtain historical maintenance data of bus lines and analyze whether there are repeated maintenance records. If there are repeated maintenance records, extract relevant maintenance resource allocation information; Adopting the optimization algorithm model, according to the extracted maintenance resource allocation information, the resource reallocation plan is calculated to determine the optimal resource allocation strategy; Generate bus line maintenance plan based on optimal resource allocation strategy, allocate maintenance resources and avoid duplication of resources; Obtain equipment operation status data and combine it with the maintenance plan to determine whether the equipment service life is affected by the maintenance. If so, adjust the maintenance frequency; By optimizing the algorithm model, the impact of the adjusted maintenance frequency on resource allocation is calculated to generate a new resource allocation plan; Update bus line maintenance plan according to the new resource allocation plan to reduce resource waste and extend equipment service life; Obtain updated maintenance plan data and analyze resource allocation efficiency. If there is an imbalance in resource allocation, optimize the algorithm model again to generate the final plan.

12. The method according to claim 1, characterized in that The operation data of the special startup combination is obtained, an operation risk judgment model is established, and the operation risk of the power grid under the left bank denominator and right bank combined denominator operation modes is identified, including: Obtain the operation data of the special startup mode, and extract relevant features based on the operation modes of the left bank denominator and the right bank combined denominator; Through the preset data processing method, the operation data is cleaned and standardized to obtain normalized data; Using the preset risk model algorithm, the standardized data is input into the model to establish a grid operation risk judgment model; If the data characteristics of the left bank denominator operation mode meet the preset threshold, it is judged to be a high-risk type; If the data characteristics of the right bank mother-combination operation mode meet the preset threshold, it is judged to be of low risk type; Based on the risk assessment results, generate risk level classification for the left bank denominator and right bank combined denominator operation modes; By classifying the risk levels output by the model, the priority of power grid operation risks is determined and the risk identification task is completed.

13. The method according to claim 1, characterized in that If there is an operational risk, the startup combination optimization algorithm is used to adjust the operation mode to ensure that the maximum power generation capacity of the plant meets the reservoir operation plan, including: Obtain reservoir operation plans and plant operation data to identify potential operation risks; Use risk identification algorithms to determine whether the current operation mode meets the requirements of the reservoir plan; If there is a risk, an optimization algorithm is used to generate a startup combination adjustment plan; Reconfigure plant and station operating parameters according to the adjustment plan to optimize power generation capacity; Use capacity matching algorithms to verify whether the power generation capacity of the plant meets the planned needs of the reservoir; If the match is successful, the adjusted operation mode will be executed to control the operation risk; Monitor operating status in real time to ensure the continued effectiveness of adjustment plans and controllability of risks.

Images