Water-wind-light joint optimization scheduling method considering operation characteristics of hydroelectric generating set

Abstract

The invention discloses a water-wind-light combined optimization scheduling method considering the operation characteristics of a hydroelectric generating set, and the method comprises the steps: synchronously obtaining wind-light-water load time sequence data according to a unified time resolution, carrying out the self-adaptive abnormal data elimination, and collecting and obtaining the static data of a hydropower station; establishing a water head-vibration area boundary discrete mapping table of each hydroelectric generating set, obtaining the upper and lower limits of the power of the vibration area in each time period by real-time water head interpolation, recording the starting and stopping events of the generating set, and counting the starting and stopping times; taking the average output of the hydroelectric generating set in each time period as a decision variable, and constructing a water-wind-light combined multi-objective optimization model; solving the multi-objective optimization model; and performing normalization and Euclidean distance calculation on the obtained solution set, and outputting a recommended scheduling scheme. According to the method, the defects that a vibration area is neglected, the water balance error is large and multi-target decision making is difficult in traditional dispatching are overcome, wind and light full consumption, unit start-stop-free and zero-forbidden-area operation are achieved, the power grid safety is improved, and the service life of a hydropower station is prolonged.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system operation and multi-energy complementary technology, and particularly relates to a water-wind-solar combined optimization scheduling method considering operation characteristics of hydroelectric generating units. BACKGROUND

[0002] With the popularization of new energy, renewable energy represented by wind energy and solar energy has developed rapidly in the world, and its share in the power system continues to rise. However, the inherent characteristics of these renewable energy power generation methods have brought unprecedented challenges to the safe, stable and economic operation of the power system.

[0003] Wind power generation and photovoltaic power generation have significant intermittency, volatility and randomness. Their output is directly affected by weather conditions (wind speed, light intensity), which is difficult to accurately predict and changes rapidly. When large-scale such power sources are connected to the power grid, they will have a profound impact on various aspects of the power system, such as power system safety and stability, power quality, prediction and scheduling complexity.

[0004] In the face of the above challenges, hydropower, especially hydropower stations with large regulating reservoirs, is considered a very valuable and ideal regulating resource in the power system. Its main advantage is that it can quickly respond and flexibly adjust. The water turbine can be quickly started and stopped, the output adjustment range is wide and fast, and it can quickly respond to changes in grid frequency and load instructions, effectively smoothing the output fluctuations of wind and solar renewable energy. Therefore, fully tapping the regulating potential of hydropower energy and achieving optimal scheduling of water, wind, and solar energy is one of the key ways to improve renewable energy consumption and ensure the safe and economic operation of the power grid. Although hydropower has excellent regulating performance, its potential has not been fully realized. In particular, traditional water power scheduling often focuses on a single economic goal. In this mode, the operation strategy of the water turbine may conflict with the need to absorb large amounts of fluctuating renewable energy. In addition, in many existing optimization scheduling models, the mathematical model of the water turbine and hydropower station is simplified, and the complexity of the operating constraints of the water turbine, especially the existence of the "vibration zone", cannot be effectively considered. When the water turbine operates at a specific output and water head combination, it may enter a region of hydraulic instability, causing the unit to vibrate violently, produce noise, and in severe cases, damage equipment, shorten the life of the unit, and even threaten the safety of the power station. Some studies suggest that the vibration zone should be divided into prohibited operation zones, vibration zones, and stable operation zones based on the degree of vibration of the unit at different loads. If the scheduling model does not accurately account for these nonlinear operating zone constraints, the scheduling plan may be difficult to implement in practice, or while it can be implemented, it may accelerate the wear and tear of the unit, leading to increased operating risks and maintenance costs. Therefore, effectively solving the optimal scheduling problem of the water, wind, and solar combined system is essentially a complex multi-objective decision-making process. It is necessary to balance and choose between multiple mutually influencing and even conflicting objectives, such as maximizing the power generation benefits of hydropower itself (such as maximizing power generation and minimizing water consumption), ensuring the safe and stable operation of the power grid (such as minimizing residual load fluctuations), and minimizing the operating costs and risks of the water turbine (such as avoiding vibration zone operation and reducing the number of starts and stops). The traditional single-objective optimization method, or the simple approach of weighting multiple objectives as a single objective, cannot effectively find a Pareto solution set that balances each objective function. Therefore, there is an urgent need for an advanced scheduling method that accurately describes the operating characteristics of the water turbine, optimizes multiple key objectives, and uses efficient multi-objective optimization algorithms. SUMMARY

[0005] To solve the above problems, the present application provides a water, wind, and solar combined optimal scheduling method considering the operating characteristics of the water turbine. By accurately modeling the operating characteristics of the water turbine and optimizing multiple objectives, the method overcomes the shortcomings of traditional scheduling, such as ignoring the vibration zone, large water balance errors, and difficulties in multi-objective decision-making, and achieves full consumption of wind and solar energy, no starts and stops of the unit, and zero forbidden zone operation, thereby improving the safety of the power grid and the service life of the hydropower station.

[0006] This invention provides a method for joint optimization scheduling of hydropower, wind power, and solar power that takes into account the operating characteristics of hydropower units. The specific technical solution is as follows: S1: Synchronously acquire time-series data of wind power, photovoltaic power, water flow and system load at a unified time resolution, perform adaptive abnormal data removal and time resolution unification processing on the time-series data, and simultaneously acquire static data of the hydropower station. S2: Establish a discrete mapping table of head-vibration zone boundary for each hydropower unit, and obtain the upper and lower limits of vibration zone power for each time period by interpolating the real-time head. At the same time, record the unit start-up and shutdown events and count the number of start-ups and shutdowns. S3: Using the average output of hydropower units in each time period as the decision variable, construct a multi-objective optimization model for hydropower, wind power and solar power, and set constraints. S4: Solve the multi-objective optimization model; S5: Based on the weighted adaptive ideal point method, the obtained solution set is normalized and Euclidean distance is calculated to output a recommended scheduling scheme.

[0007] Furthermore, in step S1, the adaptive abnormal data removal is specifically as follows: S101: Perform outlier detection on each time series data curve using a set sliding window and mark anomalies; S102: Merge consecutive outliers and generate replacement values ​​using linear interpolation of the two nearest valid points; S103: After interpolation, perform outlier recheck. If the value still exceeds the limit, replace it with the value of the adjacent valid time period.

[0008] Furthermore, the static data of the hydropower station includes the water level-reservoir capacity relationship curve, the discharge flow-tailwater level relationship curve, the NHQ curve of the unit, and the boundary data of the unit vibration zone as the water head changes.

[0009] Furthermore, in step S2, the head calculation is as follows: S201: Based on the reservoir storage volume of the previous period, refer to the water level and reservoir capacity curve to obtain the upstream water level at the beginning of this period, and use the tailwater level of the previous period to calculate the first estimated value of the water head for this period.

[0010] S202: Based on the current head estimate and unit output plan, the power generation flow rate is obtained by referring to the unit curve; the power generation flow rate is used as the outflow flow rate, and the average tailwater level for this period is obtained by referring to the tailwater level curve; Subtracting the outflow from the inflow yields the change in storage capacity over a given period. This change is then added to the initial storage capacity of the period to obtain the final storage capacity. The upstream water level at the end of the time period can be obtained by looking up the reservoir capacity curve at the end of the time period. The new head estimate is obtained by subtracting the tailwater level from the average of the upstream water level at the beginning and end of the time period. Compare the old and new head estimates. If the difference is less than the set range, record the current head, reservoir capacity and upstream and downstream water levels. Otherwise, update the head estimate and repeat step S202. S303: Use the reservoir capacity, upstream water level, and tailwater level at the end of this period as the initial conditions for the next period, and continue the calculation according to the same process until the last period ends.

[0011] Furthermore, in step S3, the water-wind-solar joint multi-objective optimization model includes at least five objective functions, which are as follows: The following parameters are considered: minimum residual load variance, maximum total power generation of the hydropower station, maximum vibration isolation depth of the unit output from the edge of the nearest vibration zone, minimum number of times the unit crosses the vibration zone, and minimum number of times the unit starts and stops.

[0012] Furthermore, in step S3, the constraints include at least: total water volume constraints, water balance constraints, unit output constraints, and reservoir water level operation constraints.

[0013] Furthermore, in step S4, the model solution process is as follows: S401: Randomly generate an initial set of scheduling schemes that satisfy the constraints. Each scheme constitutes a chromosome, and the chromosome is encoded as the time-period output sequence of all hydropower units within the scheduling period. S402: Perform binary crossover and polynomial mutation on chromosomes to obtain offspring, which are then merged with the parents to form a temporary population; S403: Time-sharing compensation for water conservation correction on an individual basis in the temporary population; S404: Perform Deb constraint judgment on the corrected individuals, eliminate infeasible solutions, then perform non-dominated sorting on the feasible solutions and associate them with the reference point, retain the front row individuals to form a new generation of population; repeat steps S402-S404 until the maximum number of generations is reached or the Pareto front remains unchanged for multiple generations, and output the final Pareto solution set.

[0014] Furthermore, in step S403, the time-sharing compensation water conservation correction is specifically as follows: For each individual in the temporary population, the reservoir capacity is calculated on a time-by-time basis according to the current output sequence. If the absolute value of the difference between the predicted final water storage and the target final water level exceeds the set threshold, the output is reduced by a soft reduction coefficient; the remaining deviation water volume after soft reduction is linearly and evenly distributed to each time period; if it does not exceed the set threshold, step S404 is executed.

[0015] Furthermore, the output force is scaled down using a soft scaling factor, as follows:

[0016] in, This is the soft scaling factor. The target final water level, This is to predict the difference between the final water storage and the target final water level.

[0017] Furthermore, in step S5, the weight vector of the five objectives is dynamically adjusted using the analytic hierarchy process or expert scoring method. After normalizing the target value to 0-1, calculate the Euclidean distance from each solution to the weighted ideal point, and take the solution with the smallest distance as the final scheduling scheme.

[0018] The beneficial effects of this invention are as follows: 1. This invention incorporates the dynamic exclusion zone of head-vibration zone into a multi-objective optimization framework, simultaneously solving five conflicting objectives: load smoothing, power generation, vibration isolation depth, number of crossings, and number of start-stop cycles, generating an executable Pareto front solution set. Compared with traditional single-objective or fixed-weight schemes, this invention smooths out renewable energy fluctuations, achieves full wind and solar power absorption, and prevents units from accidentally entering the vibration zone, extending equipment life, reducing operation and maintenance costs, and improving grid security and hydropower station lifespan.

[0019] 2. In data acquisition, this invention employs a 4-hour sliding window + 3σ / 2σ dual-layer outlier detection combined with linear interpolation and endpoint extrapolation. This enables automatic identification and repair of abnormal data in scenarios with drastic fluctuations in wind and solar power, avoiding scheduling deviations caused by bad data and providing clean and continuous input curves for subsequent optimization.

[0020] 3. This invention solves the problem that traditional fixed restricted zones cannot adapt to changes in reservoir water level by pre-establishing a discrete mapping table of head and vibration zone and using online cubic spline interpolation, so that the boundary of the restricted zone is dynamically updated with the real-time water level. This ensures that the output limit at any time corresponds completely with the actual safe zone of the unit in the head calculation. At the same time, through a closed-loop iterative mechanism of assuming head within a time period, obtaining flow rate from the curve, calculating reservoir capacity, and back-calculating head, the water volume error caused by determining the head first and then calculating the flow rate is eliminated, improving the accuracy of water level control at the end of the period and fundamentally avoiding the phenomenon of water balance hard constraints being violated.

[0021] 4. In the solution process, this invention adopts time-sharing compensation water conservation correction. Through one soft scaling + linear equal distribution compensation, the end-of-period reservoir capacity deviation is reduced to within the threshold, ensuring that NSGA-III continues to generate feasible solutions during the evolution process while shortening the calculation time. This solves the problem that multi-objective evolutionary algorithms are prone to getting stuck in infeasible regions when dealing with strong water constraints. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the integrated optimization scheduling method for water, wind and solar power.

[0023] Figure 2 This is a diagram showing the incoming water flow rate.

[0024] Figure 3 This is a schematic diagram of the simulated wind and solar power curves.

[0025] Figure 4 Schematic diagram of irregular vibration zone.

[0026] Figure 5 Schematic diagram of the vibration zone.

[0027] Figure 6 A schematic diagram of the power output curves of the power plant and the wind and solar power output curves.

[0028] Figure 7 A schematic diagram of power grid load, wind and solar power, and power plant output curves.

[0029] Figure 8 Actual output diagram of each unit.

[0030] Figure 9 Optimize the output diagrams of each unit.

[0031] Figure 10 Schematic diagram of water level operation during the optimized scheduling period. Detailed Implementation

[0032] The technical solutions in the embodiments of the present invention are clearly and completely described in the following description. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0033] In the description of the embodiments of the present invention, it should be noted that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is conventionally placed during use, or the orientation or positional relationship in which those skilled in the art conventionally understand it during use. This is only for the convenience of describing the present invention and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of the present invention. Furthermore, the terms "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0034] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0035] Example 1 Embodiment 1 of the present invention discloses a method for joint optimization scheduling of hydropower, wind power, and solar power that takes into account the operating characteristics of hydropower units, such as... Figure 1 As shown, the specific process is as follows: S1: Simultaneously acquire simulated wind power curves, simulated photovoltaic power curves, inflow curves, and system load curves at the required uniform time resolution (e.g., 5 min, 15 min). Perform adaptive outlier removal and time resolution unification processing on the time-series data. Simultaneously, acquire static data of the hydropower station; the inflow curve is as follows: Figure 2 As shown; Specifically, if different time resolutions are inconsistent, linear interpolation can be used to align them to the same time resolution.

[0036] In this embodiment, the static data of the hydropower station includes the water level-reservoir capacity relationship curve, the discharge flow-tailwater level relationship curve, the NHQ curve of the unit, and the boundary data of the unit vibration zone as the water head changes.

[0037] In a preferred embodiment, the adaptive abnormal data removal is specifically as follows: S101: Perform 3σ outlier detection on each time series data curve using a 4h sliding window. If the data points satisfy... If so, it is marked as an exception; In this embodiment, the first 4h sliding window of input data is:

[0038] Based on the data in the table above, the mean and standard deviation of each input data in this window were calculated as follows: μ1=26.77, σ1=6.17; μ2=0, σ2=0; μ3=25.23, σ3=1.30; (The wind power simulation data for time period 8 are also included.) This data was identified as an outlier.

[0039] S102: Merge consecutive outliers and generate replacement values ​​using linear interpolation of the two nearest valid points. For each outlier interval, take the left and right nearest valid points. and For all Linear interpolation is represented as follows:

[0040] In this embodiment, time period 8 is modified as follows:

[0041] If the anomalous segment is at the end of the sequence, only one-sided extrapolation is used; S103: Perform a 2σ outlier check on the interpolated new value. If it still exceeds the limit, replace the value with the value from the previous time period. This ensures that no new transitions are introduced.

[0042] S2: Establish a discrete mapping table of head-vibration zone boundary for each hydropower unit, and obtain the upper and lower limits of vibration zone power for each time period by interpolating the real-time head. At the same time, record the unit start-up and shutdown events and count the number of start-ups and shutdowns.

[0043] Each hydro-generator unit has an unstable operating range with severe vibration within a specific output range, called the vibration zone. Its upper and lower limits vary with the water head. By pre-storing the data and establishing the correspondence between the upper and lower limits of the unit's vibration zone and water head changes, the power range of the vibration zone for each time period is obtained during scheduling calculations based on the real-time water head. This allows for the implementation of operational restrictions targeting the unit's vibration zone. Figure 4 and Figure 5 As shown; Specifically, the head-vibration zone boundary discrete mapping table is obtained through on-site vibration tests or manufacturer model curves, and cubic spline interpolation is used to calculate the upper and lower limits of the vibration zone corresponding to any head in real time.

[0044] The unit start-up and shutdown refers to the process of increasing the unit output from zero to non-zero (start-up) or decreasing it from non-zero to zero (shutdown). In order to account for the impact of frequent start-ups and shutdowns on the unit's lifespan, the number of start-ups and shutdowns is statistically analyzed, but no hard limit is set on the start-up and shutdown interval. That is, it is not mandatory for the unit to wait a certain period of time after shutdown before restarting. The number of start-ups and shutdowns is minimized through optimization.

[0045] In a preferred embodiment, the head calculation is as follows: S201: Based on the reservoir storage volume of the previous period, refer to the water level and reservoir capacity curve to obtain the upstream water level at the beginning of this period, and use the tailwater level of the previous period to calculate the first estimated value of the water head for this period.

[0046] S202: Based on the current head estimate and unit output plan, the power generation flow rate is obtained by referring to the unit curve; the power generation flow rate is used as the outflow flow rate, and the average tailwater level for this period is obtained by referring to the tailwater level curve; Subtracting the outflow from the inflow yields the change in storage capacity over a given period. This change is then added to the initial storage capacity of the period to obtain the final storage capacity. The upstream water level at the end of the time period can be obtained by looking up the reservoir capacity curve at the end of the time period. The new head estimate is obtained by subtracting the tailwater level from the average of the upstream water level at the beginning and end of the time period. Compare the old and new head estimates. If the difference is less than the set range, record the current head, reservoir capacity and upstream and downstream water levels. Otherwise, update the head estimate and repeat step S202. S303: Use the reservoir capacity, upstream water level, and tailwater level at the end of this period as the initial conditions for the next period, and continue the calculation according to the same process until the last period ends.

[0047] S3: Using the average output of hydropower units in each time period as the decision variable, construct a multi-objective optimization model for hydropower, wind power and solar power, and set constraints. In this embodiment, the water-wind-solar joint multi-objective optimization model includes at least five objective functions, aiming to simultaneously optimize multiple conflicting or related objectives; Among them, the decision variable is the average output of each hydropower unit in each scheduling period t within the scheduling cycle. (in (Numbered for the hydropower units), these The decision variables of dimension 1 are the solutions of the multi-objective optimization model; Specifically, the objective function is as follows: Minimizing the mean square error of surplus load: By optimizing the scheduling of hydropower, the volatility of wind and solar power is compensated for, ensuring that the surplus load of the power grid after deducting the output of wind, solar, and hydropower is as smooth as possible. The specific calculation formula is as follows:

[0048]

[0049]

[0050] Where F is the mean square error of the grid surplus load after hydropower peak shaving, and T is the total number of time periods within the scheduling cycle. Let t be the grid load during time period t. The remaining load on the power grid after deducting the output of hydropower, wind power, and solar power during time period t. This represents the average residual load of the power grid during each time period within the dispatching period. , , These represent the output of hydropower, solar power, and wind power during time period t.

[0051] Maximizing the total power generation of the hydropower station: This involves maximizing the total power generation of the hydropower station throughout the entire scheduling cycle, while satisfying all operational constraints. The specific calculation formula is as follows:

[0052] In the formula, F represents power generation, T represents the total number of time periods within the scheduling cycle, and n(t) represents the number of generating units in operation during time period t. Let be the average output of the i-th unit during time period t. The time period is long.

[0053] The maximum vibration isolation depth occurs when the unit's output is furthest from the edge of the nearest vibration zone: that is, the load difference between the current unit's output and the edge of the nearest vibration zone is the largest. Assessing the vibration isolation depth requires simultaneously measuring the distance between the current load and the upper limit of vibration zone 1, as well as the upper and lower limits of vibration zone 2, which is a rather cumbersome process. To simplify the process, two reference values ​​can be introduced: the intermediate load of operational zone 1. and maximum load Then it is only necessary to calculate the minimum distance from the current load to one of these two points. If the current load is within the operable zone 2, then For the load during this period and The absolute value of the difference; otherwise, for and The absolute value of the difference; the specific formula is:

[0054] In the formula, To characterize the extent to which the load is far from the vibration zone, when When the minimum value is reached, the vibration damping depth is maximized.

[0055] Minimize the number of times the unit crosses the vibration zone: When the vibration zone state of the unit's output changes between two consecutive scheduling periods, it is considered that a vibration zone crossing event has occurred. To facilitate understanding of the logical process of a hydropower unit crossing the vibration zone, taking a hydropower unit with 2 vibration zones and 2 stable operating zones as an example, the operating area is divided into vibration zone 1, stable operating zone 1, vibration zone 2, and stable operating zone 2. When the unit's output is in vibration zone 1 in the current period and not in vibration zone 1 in the next period, it is considered that a vibration zone crossing has occurred. That is, when the operating zone state of the unit's output changes between two consecutive scheduling periods, it is considered that a vibration zone crossing event has occurred.

[0056] Minimize the number of unit start-ups and shutdowns: Minimizing the number of start-ups and shutdowns reduces unit operation and maintenance costs, extends equipment lifespan, and mitigates operational risks. The number of start-ups and shutdowns includes start-up events and shutdown events, specifically defined as follows: Stop the machine with zero output.

[0057] The start event is:

[0058] The downtime event was:

[0059] set up

[0060] The specific formula for minimizing the number of start-stop cycles is: .

[0061] The model must satisfy a series of physical and operational constraints to ensure the feasibility and safety of the scheduling scheme. The main constraints include: total water volume constraint, water balance constraint, hydropower unit output constraint, and reservoir operating water level constraint. For this embodiment, the total water volume constraint is 195.7 million cubic meters of usable water calculated from the initial and final water levels and the inflow. The output of each unit is 0-700MW, and the minimum operating water level of the reservoir is the dead water level of 540m, and the maximum is the normal storage water level of 600m. Specifically, the constraints are as follows: Total water volume constraint: Within a complete scheduling cycle T, the total outflow of water from a single reservoir must not exceed the total planned water consumption for the scheduling period. In actual scheduling, it is generally more convenient to provide the planned end-of-period water level. In this case, the final reservoir storage volume can be obtained from the water level-storage curve using the end-of-period water level. The total water consumption can be obtained by subtracting the final reservoir storage volume from the initial storage volume and adding the planned inflow volume. The specific formula is as follows: .

[0062] Water balance constraints:

[0063] in, , These represent the initial and final reservoir water storage volumes for time period t. , , These represent the average inflow, outflow, and wastewater discharge during time period t.

[0064] Unit output constraints:

[0065] in, , These are the minimum and maximum outputs of unit i during time period t, respectively.

[0066] Reservoir water level operation constraints:

[0067] in, , These represent the minimum and maximum allowable water levels upstream of the reservoir at the beginning of time period t.

[0068] S4: The NSGA-III algorithm, which introduces a time-sharing water conservation operator, is used to solve the multi-objective optimization model to obtain the Pareto front solution set.

[0069] As a preferred embodiment, the model solution process is as follows: S401: Randomly generate an initial set of scheduling schemes that satisfy the constraints. Each scheme constitutes a chromosome, and the chromosome is encoded as the time-period output sequence of all hydropower units within the scheduling period. Specifically, during the generation process, each initial solution must satisfy hard constraints, especially the output constraints of the hydropower units.

[0070] S402: Perform binary crossover and polynomial mutation on chromosomes to obtain offspring, which are then merged with the parents to form a temporary population; S403: Time-sharing compensation for water conservation correction on an individual basis in the temporary population; The time-sharing compensation water conservation correction is as follows: For each individual in the temporary population, according to the current output sequence Calculate reservoir capacity by time period ; If the final water storage volume is predicted With the target final water level The absolute value of the difference satisfies The output force is then scaled using a soft scaling factor, as follows:

[0071] in, This is the soft scaling factor. The target final water level, To predict the difference between the final water storage and the target final water level; Recalculate the predicted water level and target water level after the soft release reduction. The adjustable output is evenly distributed to each time period, and the evenly distributed water volume is linearly approximated to the evenly distributed output for time-period compensation. If the set threshold is not exceeded, proceed to step S404.

[0072] The time-sharing compensation water conservation operator, through soft scaling coefficients and hourly linear compensation, ensures that the error between the reservoir capacity at the end of the scheduling period and the target reservoir capacity is less than a set threshold, thereby guaranteeing that the water balance is strictly met.

[0073] S404: Perform Deb constraint judgment on the corrected individuals, eliminate infeasible solutions, then perform non-dominated sorting on the feasible solutions and associate them with the reference point, retain the front row individuals to form a new generation of population; repeat steps S402-S404 until the maximum number of generations is reached or the Pareto front remains unchanged for multiple generations, and output the final Pareto solution set.

[0074] S5: Based on the weighted adaptive ideal point method, normalize the obtained Pareto front solution set and calculate the Euclidean distance to output a recommended scheduling scheme.

[0075] As a preferred embodiment, the weight vector of the five objectives is dynamically adjusted using the analytic hierarchy process or expert scoring method. After normalizing the target value to 0-1, calculate the Euclidean distance from each solution to the weighted ideal point, and take the solution with the smallest distance as the final scheduling scheme.

[0076] Based on the above method, this embodiment uses the initial water level of 567.158m and the final water level of 567.61m on a certain day during the dry season of a power station as input, optimizes the number of generating units to 6, and adopts the "water-determined power generation" scheduling rule. The inflow rate is based on data from the cascade dispatching system to which the reservoir belongs, such as... Figure 2 As shown, since the actual data timescale is 1 hour, the inflow rate at the 15-minute scale was obtained by interpolation. The flow rate process exhibits a sawtooth pattern, mainly because the inflow rate is calculated by reverse calculation, which has no impact on the verification of the method of this invention. The wind and solar data were simulated using wind and solar planning data near the hydropower station (wind power installed capacity 876MW, solar power installed capacity 3670MW), as shown below. Figure 3 As shown in the figure. Based on the above method, a daily scheduling optimization scheme is obtained, which realizes the coordinated allocation of wind power, photovoltaic and hydropower output, effectively mitigates the impact of renewable energy fluctuations on the power grid, and ensures the reservoir operation constraints.

[0077] The main features of the scheduling scheme are as follows: (see attached) Figure 6 As shown, during periods of high renewable energy output (such as the midday peak of photovoltaic power), hydropower output is appropriately reduced to make way for wind and solar power generation—the generating units operate at their minimum stable output, maximizing the use of clean energy for power generation; such as Figure 7 During peak load periods in the morning and evening, or when wind and solar power output is insufficient, hydropower stations increase their output to make up for the shortfall—the unit output can be adjusted upwards within a safe range to meet the system's peak load demand. This scheduling significantly reduces the variation in the residual load of the power grid after deducting wind, solar, and hydropower, and the peak-to-valley difference tends to be smoother.

[0078] The scheduling results show that by assigning weights of 0.3, 0.1, 0.3, 0.2, and 0.1 to the residual load variance, maximum power generation, vibration isolation depth, number of vibration zone crossings, and number of start-stop cycles, respectively, the scheme with the minimum Euclidean distance from the ideal point in the Pareto front can be obtained, such as... Figure 8 , Figure 9 The right bank peak-shaving generating units experienced zero start-ups and shutdowns, significantly reducing the frequency of unit start-ups and shutdowns and operational wear; the hydropower units experienced 25 vibration zone crossing events, consistently operating within stable sections far from the vibration zone; and the reservoir water level trajectory remained consistently within permissible limits. Figure 10The final water level reached 567.611 meters, with an error of 0.01 meters, meeting the requirements for water balance and final water level control. Therefore, the optimized scheduling scheme in this embodiment fully satisfies the constraints and has achieved good results in improving renewable energy utilization, smoothing wind and solar power output fluctuations, and ensuring the safe operation of generating units. The actual scheduling indicators were calculated using actual data according to the definitions of each indicator. Due to the large grid load, all wind and solar power were absorbed during the actual scheduling.

[0079] This invention is not limited to the specific embodiments described above. The invention extends to any new feature or combination disclosed in this specification, as well as any new method or process step or combination disclosed herein.

Claims

1. A method for joint optimization scheduling of hydropower, wind power, and solar power considering the operating characteristics of hydropower units, characterized in that, include: S1: Synchronously acquire time-series data of wind power, photovoltaic power, water flow and system load at a unified time resolution, perform adaptive abnormal data removal and time resolution unification processing on the time-series data, and simultaneously acquire static data of the hydropower station. S2: Establish a discrete mapping table of head-vibration zone boundary for each hydropower unit, and obtain the upper and lower limits of vibration zone power for each time period by interpolating the real-time head. At the same time, record the unit start-up and shutdown events and count the number of start-ups and shutdowns. S3: Using the average output of hydropower units in each time period as the decision variable, construct a multi-objective optimization model for hydropower, wind power and solar power, and set constraints. S4: Solve the multi-objective optimization model; S5: Based on the weighted adaptive ideal point method, the obtained solution set is normalized and Euclidean distance is calculated to output a recommended scheduling scheme.

2. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, In step S1, the adaptive abnormal data removal is specifically as follows: S101: Perform outlier detection on each time series data curve using a set sliding window and mark anomalies; S102: Merge consecutive outliers and generate replacement values ​​using linear interpolation of the two nearest valid points; S103: After interpolation, perform outlier recheck. If the value still exceeds the limit, replace it with the value of the adjacent valid time period.

3. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, The static data of the hydropower station includes the water level-reservoir capacity relationship curve, the discharge flow-tailwater level relationship curve, the NHQ curve of the generating unit, and the boundary data of the unit vibration zone as the water head changes.

4. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, In step S2, the head calculation is as follows: S201: Based on the reservoir storage volume in the previous period, look up the water level and reservoir capacity curve to obtain the upstream water level at the beginning of this period, and use the tailwater level in the previous period to calculate the first estimated value of the water head in this period. S202: Based on the current head estimate and unit output plan, the power generation flow rate is obtained by referring to the unit curve; the power generation flow rate is used as the outflow flow rate, and the average tailwater level for this period is obtained by referring to the tailwater level curve; Subtracting the outflow from the inflow yields the change in storage capacity over a given period. This change is then added to the initial storage capacity of the period to obtain the final storage capacity. The upstream water level at the end of the time period can be obtained by looking up the reservoir capacity curve at the end of the time period. The new head estimate is obtained by subtracting the tailwater level from the average of the upstream water level at the beginning and end of the time period. Compare the old and new head estimates. If the difference is less than the set range, record the current head, reservoir capacity and upstream and downstream water levels. Otherwise, update the head estimate and repeat step S202. S303: Use the reservoir capacity, upstream water level, and tailwater level at the end of this period as the initial conditions for the next period, and continue the calculation until the last period ends.

5. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, In step S3, the water-wind-solar joint multi-objective optimization model includes at least five objective functions, which are as follows: The following parameters are considered: minimum residual load variance, maximum total power generation of the hydropower station, maximum vibration isolation depth of the unit output from the edge of the nearest vibration zone, minimum number of times the unit crosses the vibration zone, and minimum number of times the unit starts and stops.

6. The method for joint optimization scheduling of water, wind, and solar power according to claim 1, characterized in that, In step S3, the constraints include at least: total water volume constraints, water balance constraints, unit output constraints, and reservoir water level operation constraints.

7. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, In step S4, the model solution process is as follows: S401: Randomly generate an initial set of scheduling schemes that satisfy the constraints. Each scheme constitutes a chromosome, and the chromosome is encoded as the time-period output sequence of all hydropower units within the scheduling period. S402: Perform binary crossover and polynomial mutation on chromosomes to obtain offspring, which are then merged with the parents to form a temporary population; S403: Time-sharing compensation for water conservation correction on an individual basis in the temporary population; S404: Perform Deb constraint judgment on the corrected individuals, eliminate infeasible solutions, then perform non-dominated sorting on the feasible solutions and associate them with the reference point, retain the front row individuals to form a new generation of population; repeat steps S402-S404 until the maximum number of generations is reached or the Pareto front remains unchanged for multiple generations, and output the final Pareto solution set.

8. The water-wind-solar joint optimization scheduling method according to claim 7, characterized in that, In step S403, the time-sharing compensation water conservation correction is specifically as follows: For each individual in the temporary population, the reservoir capacity is calculated on a time-by-time basis according to the current output sequence. If the absolute value of the difference between the predicted final water storage and the target final water level exceeds the set threshold, the output will be reduced using a soft reduction coefficient. The remaining deviation water volume after the soft release is linearly and evenly distributed to each time period; If the set threshold is not exceeded, proceed to step S404.

9. The water-wind-solar joint optimization scheduling method according to claim 8, characterized in that, The output force is scaled using a soft scaling factor, as follows: in, This is the soft scaling factor. The target final water level, This is to predict the difference between the final water storage and the target final water level.

10. The water-wind-solar joint optimization scheduling method according to claim 1, characterized in that, In step S5, the weight vector of the five objectives is dynamically adjusted using the analytic hierarchy process or expert scoring method. After normalizing the target value to 0-1, calculate the Euclidean distance from each solution to the weighted ideal point, and take the solution with the smallest distance as the final scheduling scheme.

Images