Complex watershed series-parallel hydropower station group installation scheme determination method

Abstract

The invention discloses a complex watershed series-parallel hydropower station group installation scheme determination method, which comprises the following steps of: determining the number of hydropower stations required by a research area, dividing a plurality of hydropower station units, and clustering the plurality of hydropower station units by adopting a depth embedding clustering algorithm to form a plurality of similar watershed units; calculating the number of hydropower stations and the primary installed capacity in each similar drainage basin unit; a decision-making model based on transform is adopted, the hydropower station and the initial installed capacity are used as input, and the ultimate installed capacity of the hydropower station is obtained; constructing a hybrid hydropower station group optimization scheduling model, solving the model by taking the ultimate hydropower station installed capacity as an input to obtain the maximum generating capacity, drawing a relation curve of the generating capacity and the cost by adopting a gradual element reduction method, identifying a critical point of the relation curve, and taking the critical point as the most economic point to obtain an optimal hydropower station installation scheme. According to the invention, low power generation efficiency and resource waste caused by improper layout are reduced, and the economic benefit and operation stability of the hydropower station are improved.

Description

Technical Field

[0001] The invention relates to a method for determining an installation plan for a group of hybrid hydropower stations in a complex river basin. Background Art

[0002] As the global demand for clean energy continues to rise, the development and utilization of hydropower, as an important renewable energy source, has far-reaching significance for energy structure optimization and sustainable development.

[0003] For a group of hybrid hydropower stations in a complex river basin, accurate determination of installed capacity is the core link to achieve efficient and stable power generation and improve energy efficiency. Traditional methods for determining installed capacity often have many limitations when faced with complex geological conditions, changeable meteorological factors, and diverse spatial influencing factors of hydropower stations. In terms of geological conditions, traditional methods often use simple geological survey methods, which are difficult to fully detect complex geological structures. Conditions such as faults, folds, and special rock and soil distribution may be ignored or not accurately grasped, resulting in the inability to fully consider the impact of geological stability on engineering safety and operating costs when selecting and constructing hydropower stations. The collection and analysis of meteorological data such as precipitation, evaporation, and temperature is relatively simple, and there is a lack of long-term, continuous and high-resolution monitoring. When calculating hydropower resources, the impact of dynamic changes in meteorological factors on water resources cannot be fully considered; the spatial influencing factors of hydropower stations are complex and diverse, covering topography, load requirements, and energy types. Traditional methods find it difficult to comprehensively consider the relationship and synergy between these factors, and cannot fully consider the impact of terrain on the distribution and utilization of hydropower. In addition, there is a lack of accurate prediction of the dynamic changes in electricity market demand, which makes the installed capacity of hydropower stations mismatch with actual electricity demand, affecting the stability and economy of energy supply.

[0004] The present invention proposes a method for determining the installation plan of a group of hybrid hydropower stations in a complex river basin, which solves the above-mentioned problems currently existing, provides a more reliable and scientific basis for the planning and construction of hydropower stations, and promotes the efficient development of the hydropower industry. Summary of the invention

[0005] Purpose of the invention: To provide a method for determining the installation plan of a group of hybrid hydropower stations in a complex river basin, so as to solve the above-mentioned problems existing in the prior art.

[0006] Technical solution: 1. A method for determining the installation plan of a complex river basin hybrid hydropower station group, characterized by comprising the following steps:

[0007] Step S1, determine the number of hydropower stations required in the study area, divide several hydropower station units, compress and cluster them, and form several similar watershed units;

[0008] Step S2, calculating the number of hydropower stations and primary installed capacity in each similar watershed unit;

[0009] Step S3: using a transformer-based decision model, taking the hydropower station and the initial installed capacity as input, to obtain the ultimate installed capacity of the hydropower station;

[0010] Step S4, input the ultimate installed capacity of the hydropower station into the pre-built hybrid hydropower station group optimization scheduling model, use the fish algorithm to solve and obtain the maximum power generation and use it as the upper bound, deduct the hydropower station cost influencing factors in turn and calculate the corresponding power generation, draw the power generation-cost relationship curve, identify the critical point of the curve as the most economical point, and determine the optimal installed capacity plan of the hydropower station.

[0011] According to one aspect of the present application, step S4 is specifically as follows: inputting the ultimate installed capacity of the hydropower station into the pre-built hybrid hydropower station group optimization scheduling model, using the fish algorithm to solve the model, obtaining the maximum power generation, and using it as the upper bound, then subtracting the hydropower station cost influencing factors in turn, and calculating the corresponding power generation, drawing the relationship curve between power generation and cost, identifying the critical point of the relationship curve and using it as the most economical point, and obtaining the optimal installed capacity plan for the hydropower station.

[0012] According to one aspect of the present application, step S1 further comprises:

[0013] Step S11, collect the data of the study area, including the hydrological data of the study area and the spatial unit influencing factors of the hydropower station, wherein the hydrological data of the study area include: basin area, river length, natural river runoff, river network density, water system shape, land use mode and soil type; the spatial influencing factors of the hydropower station include: precipitation depth, greening degree, evaporation, evaporation area, geological conditions, topography, load requirements, energy type, natural water inflow and regulation cycle of the hydropower station;

[0014] Step S12: determining the number of hydropower stations required in the study area based on the data of the study area, and dividing the determined hydropower stations into a number of hydropower station units based on the hydropower station spatial unit influencing factors;

[0015] Step S13: construct an autoencoder that can optimize the objective function, and use several hydropower station spatial unit impact factors as input, compress and embed several hydropower station unit impact factors into a low-dimensional feature space, and then use the DBSCAN algorithm to cluster them to form several similar basin units.

[0016] According to one aspect of the present application, step S13 is further:

[0017] Step S13a, constructing an optimizable objective function autoencoder based on the division of the impact factors of the hydropower station spatial unit, including: an encoder and a decoder, setting the domain radius and the minimum number of points;

[0018] Step S13b, using an encoder to add a feature clustering layer based on the hydropower station spatial unit impact factor at the output, and compressing the hydropower station spatial unit impact factor and embedding it into a low-dimensional feature space;

[0019] Step S13c, for each data point in the low-dimensional feature space, calculate the number of points in the domain and determine whether it is greater than the minimum number of points, and take the point greater than the minimum number of points as the core point;

[0020] Step S13d, starting from a core point, all points within the core point area are added to the same cluster. When a point within the area is also a core point, the cluster is expanded with the point as the starting point until it cannot be expanded further, and several cluster centers are obtained;

[0021] Step S13e, taking several cluster centers as the objective function, optimizing the objective function through back propagation and the autoencoder model, and obtaining several optimized cluster centers to divide the study area into several similar watershed units.

[0022] According to one aspect of the present application, step S2 further comprises:

[0023] Step S21, reading and calculating the number of hydropower stations in each similar watershed unit based on the watershed area, river length, natural river runoff, river network density, water system shape, land use pattern, soil type, temperature and sunshine time of each similar watershed unit;

[0024] Step S22, respectively calculate the maximum working capacity, spare capacity and redundant capacity of the hydropower station in each similar watershed unit, and obtain the primary installed capacity of each hydropower station, which is the sum of the maximum working capacity, spare capacity and redundant capacity of the hydropower station.

[0025] According to one aspect of the present application, step S22 is further:

[0026] Step S22a, collect historical data of the hydropower station to obtain the average flow in the dry year, draw a relationship curve between the reservoir water level and storage capacity, draw a relationship curve between the reservoir storage capacity and the downstream flow, calculate the storage capacity between the dead water level and the normal water level, that is, the beneficial storage capacity of the hydropower station, use a trial algorithm to calculate the water storage period, water supply period and power generation reference flow of the hydropower station, calculate the monthly average output of the hydropower station, that is, the guaranteed output of the hydropower station, calculate the guaranteed electric energy of the hydropower station based on the guaranteed output of the hydropower station and the water supply period of the reservoir, draw a relationship curve between the working capacity and the power generation of the hydropower station, and the working capacity corresponding to the point found on the relationship curve with the guaranteed electric energy as the coordinate is the maximum working capacity of the hydropower station;

[0027] Step S22b, calculating the load reserve capacity and the accident reserve capacity based on the maximum load of the system, the reserve capacity of the hydropower station is the sum of the load reserve capacity and the accident reserve capacity;

[0028] Step S22c, based on the water inflow in normal water years and flood water years, calculate the output of the hydropower station on a monthly basis, draw a relationship curve between the abandoned water output and the multi-year average continuous output time, calculate the economic annual utilization hours, use the economic annual utilization hours as the coordinate value, and locate on the relationship curve between the abandoned water output and the multi-year average continuous output time. The abandoned water output value obtained is the repeated capacity;

[0029] Step S22d, summing up the maximum working capacity of the hydropower station, the standby capacity of the hydropower station and the redundant capacity of the hydropower station to obtain the primary installed capacity of the hydropower station.

[0030] According to one aspect of the present application, step S3 further comprises:

[0031] Step S31: construct a decision model based on heuristic rules based on Transformer, and construct the connection rules between the influencing factors of the installed capacity of each part of the hydropower station and its installed capacity;

[0032] Step S32, taking the spatial unit impact factor of the hydropower station in each similar watershed unit as the original data set, taking the first 80% of the original data set as the training set, and the last 20% as the verification set;

[0033] Step S33, input the training set into the decision model, compare the obtained result with the primary installed capacity of the hydropower station and adjust the decision model parameter weights based on the comparison result;

[0034] Step S34, extracting the validation set and inputting it into the trained decision model, solving the model to obtain the simulated installed capacity of the hydropower station and comparing it with the installed capacity of the hydropower station, and determining the accuracy of the decision model based on the comparison result, optimizing the decision model when the accuracy of the decision model is less than a threshold, repeatedly iterating until the accuracy of the decision model is greater than the threshold, and obtaining the optimized decision model;

[0035] Step S35, extract the influencing factors of the hydropower station spatial unit and input them into the optimized decision model, solve the model to obtain the ultimate installed capacity of the hydropower station.

[0036] According to one aspect of the present application, step S4 is further:

[0037] Step S41, constructing an optimal dispatching model for a hybrid hydropower station group, wherein the objective function is to maximize power generation, the decision variable is the downstream flow rate, and the constraints include: water balance of each reservoir, upper and lower limit water level constraints of each reservoir, discharge capacity of each reservoir discharge gate, downstream safety discharge, flood discharge gate operation constraints and non-negative constraints, and taking the installed capacity of the hydropower station as input, using the fish algorithm to solve the model and obtain the maximum power generation;

[0038] Step S42: collect the cost influencing factors of the hydropower station and perform sensitivity calculation, and use the stepwise reduction analysis method to calculate the power generation and cost of the power station combination after each reduction based on the sensitivity of the cost influencing factors of each hydropower station;

[0039] Step S43, draw a relationship curve between power generation and construction cost, and use piecewise linear fitting method to identify the critical point of the relationship curve as the most economical point. The corresponding combination of hydropower stations is the optimal installation plan of the hydropower station.

[0040] According to one aspect of the present application, step S41 is further:

[0041] Step S41a, setting the population size, the maximum number of iterations and the maximum moving step length of the remora, initializing individual positions, randomly generating candidate solutions, and selecting the current candidate solution as the optimal solution;

[0042] Step S41b, adjusting the candidate solution position based on the host position, inputting each individual into the hybrid hydropower station group optimization scheduling model, and calculating the corresponding total power generation, which is the fitness value of the individual;

[0043] Step S41c: for each individual, based on the fitness value of the individual, select a better individual as the follow-up object, update the current optimal solution and the candidate solution position, and make a judgment. When the fitness of the candidate solution is better, update the global optimal solution;

[0044] Step S41d, repeatedly iterate until the maximum number of iterations is reached, obtain the scheduling plan corresponding to the individual with the maximum fitness, and calculate the maximum power generation.

[0045] According to one aspect of the present application, step S42 is further:

[0046] Step S42a, collecting data on factors affecting the cost of a hydropower station, and constructing a mixed hydropower station cost function, wherein the factors affecting the cost of a hydropower station include: upstream and downstream head difference, water resource stability, geological conditions of the hydropower station, transmission line model, substation equipment model, concrete pouring method, generator set model, and turbine set model;

[0047] Step S42b, using the local perturbation method to perform sensitivity analysis on the factors affecting the cost of the hydropower station, obtain the sensitivity of the factors affecting the cost of each hydropower station, and generate the cost of each hydropower station as an initial solution based on the power generation of the hydropower station;

[0048] Step S42c, select the solution space and gradually reduce the dimensions of the hydropower station cost influencing factors in order from small to large according to the sensitivity of the hydropower station cost influencing factors to obtain the power station combination after each reduction, calculate the power generation and cost of the power station combination after the reduction, and iterate repeatedly until the maximum number of iterations.

[0049] According to one aspect of the present application, step S43 is further:

[0050] Step S43a, with the construction cost as the horizontal coordinate and the power generation as the vertical coordinate, the power generation after each reduction and the corresponding construction cost are plotted in the coordinate system, and each point is connected to obtain a power generation-construction cost relationship curve;

[0051] Step S43b, using a piecewise linear fitting method to divide the power generation-cost relationship curve into several segments, each segment is fitted with a straight line, and the slope of each straight line is calculated;

[0052] Step S43c, calculate the change in the slope of the two straight lines before and after each data point in turn, take the data point corresponding to the maximum change as the critical point, and the hydropower station combination corresponding to the critical point is the optimal installation plan of the hydropower station.

[0053] Beneficial effects: The adoption of a method for determining the installed capacity scheme of a complex-basin hybrid hydropower station group provides a more reliable and scientific basis for the planning, construction and operation of hydropower stations. It has broad application prospects and is expected to play an important role in actual projects, promoting the hydropower industry to develop in an efficient and sustainable direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of the present invention.

[0055] Figure 2 It is a flow chart of step S1 of the present invention.

[0056] Figure 3 It is a flow chart of step S2 of the present invention.

[0057] Figure 4 It is a flow chart of step S3 of the present invention.

[0058] Figure 5 It is a flow chart of step S4 of the present invention. DETAILED DESCRIPTION

[0059] like Figure 1According to one aspect of the present application, a method for determining a complex basin hybrid hydropower station installation plan is provided, which is characterized by comprising the following steps:

[0060] Step S1, determine the number of hydropower stations required in the study area, and divide it into several hydropower station units, use a pre-built optimizable objective function autoencoder to compress and embed several hydropower station units into a low-dimensional feature space, and then use a clustering algorithm to cluster them to form several similar basin units;

[0061] Step S2, calculating the number of hydropower stations and primary installed capacity in each similar watershed unit;

[0062] Step S3: using a transformer-based decision model, taking the hydropower station and the initial installed capacity as input, to obtain the ultimate installed capacity of the hydropower station;

[0063] Step S4, input the ultimate installed capacity of the hydropower station into the pre-built hybrid hydropower station group optimization scheduling model, use the fish algorithm to solve the model, obtain the maximum power generation, and use it as the upper bound, then subtract the hydropower station cost influencing factors in turn, calculate the corresponding power generation, draw the relationship curve between power generation and cost, identify the critical point of the relationship curve and use it as the most economical point, and obtain the optimal installed capacity plan of the hydropower station.

[0064] like Figure 2 As shown, according to one aspect of the present application, the step S1 is further:

[0065] Step S11, collect the data of the study area, including the hydrological data of the study area and the spatial unit influencing factors of the hydropower station, wherein the hydrological data of the study area include: basin area, river length, natural river runoff, river network density, water system shape, land use mode and soil type; the spatial influencing factors of the hydropower station include: precipitation depth, greening degree, evaporation, evaporation area, geological conditions, topography, load requirements, energy type, natural water inflow and regulation cycle of the hydropower station;

[0066] Step S12: determining the number of hydropower stations required in the study area based on the data of the study area, and dividing the determined hydropower stations into a number of hydropower station units based on the hydropower station spatial unit influencing factors;

[0067] Step S13: construct an autoencoder that can optimize the objective function, and use several hydropower station spatial unit impact factors as input, compress and embed several hydropower station unit impact factors into a low-dimensional feature space, and then use the DBSCAN algorithm to cluster them to form several similar basin units.

[0068] Comprehensively collect geological data of the study area, among which the basin area is directly related to the amount of water resources collected, and a larger basin area usually means more abundant hydropower resources; the length of the river affects the distribution and utilization of hydropower; the natural river runoff is a key indicator to measure the abundance of water resources, which directly determines the power generation potential of the hydropower station; the river network density reflects the level of development of the water system and affects the layout of hydropower development; the shape of the water system will affect the flow velocity and flow distribution of the water flow, and thus affect the site selection and construction of the hydropower station; land use methods such as forest land, cultivated land, construction land, etc. will have different degrees of impact on surface runoff; soil type is related to soil erosion, which indirectly affects the water quality and quantity of the river.

[0069] At the same time, spatial influencing factors of hydropower stations are collected. The depth of precipitation directly affects the replenishment of river water; the greening degree affects the circulation of water resources by affecting evaporation and surface runoff; the evaporation amount and evaporation area affect the loss of water resources; the geological conditions determine the feasibility and safety of hydropower station construction; the topography affects the site selection and construction difficulty of the hydropower station; the load requirement is determined according to the electricity demand of the local and surrounding areas to ensure that the power generation matches the demand; the energy type is related to the diversification and complementarity of the energy structure; the natural water inflow is the basis for the power generation of the hydropower station; the regulation cycle of the hydropower station affects the stability of the power supply.

[0070] Compared with the traditional method that only focuses on a few key factors, the above method in this embodiment effectively solves the problem of unreasonable planning caused by incomplete data, avoids engineering risks or waste of resources caused by ignoring certain important factors, and provides a solid data foundation for subsequent analysis and decision-making. It can fully consider the various complex factors that affect the construction and operation of hydropower stations, and ensure the scientificity and rationality of the planning.

[0071] According to one aspect of the present application, step S13 is further:

[0072] Step S13a, constructing an optimizable objective function autoencoder based on the division of the impact factors of the hydropower station spatial unit, including: an encoder and a decoder, setting the domain radius and the minimum number of points;

[0073] Step S13b, using an encoder to add a feature clustering layer based on the hydropower station spatial unit impact factor at the output, and compressing the hydropower station spatial unit impact factor and embedding it into a low-dimensional feature space;

[0074] Step S13c, for each data point in the low-dimensional feature space, calculate the number of points in the domain and determine whether it is greater than the minimum number of points, and take the point greater than the minimum number of points as the core point;

[0075] Step S13d, starting from a core point, all points within the core point area are added to the same cluster. When a point within the area is also a core point, the cluster is expanded with the point as the starting point until it cannot be expanded further, and several cluster centers are obtained;

[0076] Step S13e, taking several cluster centers as the objective function, optimizing the objective function through back propagation and the autoencoder model, and obtaining several optimized cluster centers to divide the study area into several similar watershed units.

[0077] In this embodiment, a deep embedding clustering algorithm is used to cluster several hydropower station units. The algorithm constructs an optimizable objective function autoencoder based on the division of the influencing factors of the hydropower station spatial units, including an encoder and a decoder. The encoder maps the high-dimensional hydropower station data to a low-dimensional feature space and extracts key features. The decoder is responsible for reconstructing the low-dimensional features into original data. By continuously optimizing the objective function, the autoencoder can better learn the feature representation of the data. Then, the DBSCAN algorithm is used to cluster in the low-dimensional feature space to obtain several clustering centers. The DEC classification is combined to optimize the objective function. The clustering results are further improved by optimizing the objective function. The obtained clustering centers are regarded as targets and optimized through back propagation and the autoencoder model.

[0078] Through this cluster analysis method, river basins with similar geological, meteorological, hydropower and other characteristics can be grouped together, which facilitates the subsequent refined and differentiated analysis and planning of different types of river basin units, improves the pertinence and effectiveness of planning, and solves the problem of traditional methods using unified standards to plan complex river basins and ignoring the differences between river basins, making the planning more in line with actual conditions.

[0079] like Figure 3 As shown, according to one aspect of the present application, the step S2 is further:

[0080] Step S21, calculating the number of hydropower stations in each similar watershed unit based on the watershed area, river length, natural river runoff, river network density, water system shape, land use pattern, soil type, temperature and sunshine time of each similar watershed unit;

[0081] Taking multiple factors into consideration comprehensively can make the layout of the number of hydropower stations in each river basin unit more scientific and reasonable, give full play to the advantages of water resources in each river basin, improve energy utilization efficiency, effectively avoid the problem of unreasonable layout of hydropower stations in similar methods, and reduce low power generation efficiency and resource waste caused by improper layout.

[0082] In a certain embodiment, specifically:

[0083] A study area with a drainage area of ​​5000km 2 , the river length is 200km, the natural river runoff is 50m 3 / s, the river network density is 0.5km / km 2 The river system is in the shape of a dendrite, which is relatively regular, with a distinct main river channel and evenly distributed tributaries. The land use is 70% forest land, 20% farmland, and 10% town and other land. The land suitable for the construction of hydropower stations accounts for about 5%, or 250km 2 , the soil types are mainly clay and loam, the geological conditions are stable, and suitable for engineering construction;

[0084] The average drop in the study area is 200m according to the topographic data. The theoretical reserves of hydropower resources are calculated using the formula E=9.81QH: E=9.81×50×200=98100kW;

[0085] Considering that some areas are difficult to develop due to factors such as ecological protection and construction costs, assuming that the development coefficient is 0.4, the amount of water resources that can be developed is: E 可开发 =9.81×50×200=98100 kW;

[0086] Based on the river network density and water system shape, assuming that every 100km2 can be divided into a relatively independent hydropower development unit, the entire basin can be divided into 5000÷100=50 development units. However, considering the land use pattern, the area suitable for the construction of hydropower stations can be divided into 250÷100=2.5 ​​development units. Since there cannot be half a development unit, it can actually be regarded as 3 development units;

[0087] According to experience and similar engineering cases, under the geological conditions and river scale, each development unit is suitable for building a hydropower station with an installed capacity of about 15,000 kW. The total number of hydropower stations that can be built in the basin is: 39,240 ÷ 15,000 ≈ 2.62, rounded down to 2.

[0088] Step S22, respectively calculate the maximum working capacity, spare capacity and redundant capacity of the hydropower station in each similar watershed unit, and obtain the primary installed capacity of each hydropower station, which is the sum of the maximum working capacity, spare capacity and redundant capacity of the hydropower station.

[0089] According to one aspect of the present application, step S22 is further:

[0090] Step S22a: Collect historical data of the hydropower station to obtain the average flow rate in dry years, draw the relationship curve between reservoir water level and storage capacity, draw the relationship curve between reservoir storage capacity and discharge flow rate, calculate the storage capacity between the dead water level and the normal water level, which is the useful storage capacity of the hydropower station, use the trial method to calculate the impounding period, supply period and power generation diversion flow rate of the hydropower station, calculate the monthly average output of the hydropower station, which is the guaranteed output of the hydropower station, calculate the guaranteed electric energy of the hydropower station based on the guaranteed output of the hydropower station and the duration of the supply period of the reservoir, draw the relationship curve between the working capacity and the power generation of the hydropower station, and the working capacity corresponding to the point found on the relationship curve with the guaranteed electric energy as the coordinate is the maximum working capacity of the hydropower station;

[0091] In a certain embodiment, specifically:

[0092] By checking the original data, obtain the average flow rate in dry years; based on the basic data of the hydropower station, draw the relationship curve between reservoir water level and storage capacity, the relationship curve between reservoir storage capacity and discharge flow rate, and find the storage capacity between the dead water level and the normal water level, which is the useful storage capacity of the reservoir; use the trial method to find the impounding period, supply period and power generation diversion flow rate of the reservoir; calculate the monthly average output of the hydropower station, which is the guaranteed output of the hydropower station;

[0093] Calculate the guaranteed electric energy of the hydropower station

[0094] Eguarantee = Nguarantee * 730 * Tsupply

[0095] Where: Nguarantee is the guaranteed output of the hydropower station; Tsupply is the duration of the supply period of the reservoir, in months

[0096] For the working capacities N1, N2, N3, N4, N5 of 5 hydropower stations, calculate the power generation during the supply period for each scheme to obtain E1, E2, E3, E4, E5, draw the N-E curve, and find the point M (Nworking, Eguarantee) on the curve. Nworking is the maximum working capacity of the hydropower station.

[0097] Step S22b: Calculate the load reserve capacity and accident reserve capacity based on the system's maximum load. The reserve capacity of the hydropower station is the sum of the load reserve capacity and the accident reserve capacity;

[0098] The reserve capacity of the hydropower station consists of two parts;

[0099] That is: Nreserve = Nload + Naccident

[0100] Where, Nreserve is the reserve capacity of the hydropower station; Nload is the load reserve capacity, which is 5% of the system's maximum load; Naccident is the accident reserve capacity, which is 10% of the system's maximum load;

[0101] Step S22c, based on the water inflow conditions in normal water years and flood water years, calculate the output of the hydropower station on a monthly basis, draw a relationship curve between the abandoned water output and the multi-year average continuous output time, calculate the economic annual utilization hours, use the economic annual utilization hours as the coordinate value, and locate on the relationship curve between the abandoned water output and the multi-year average continuous output time. The abandoned water output value obtained is the repeated capacity.

[0102] In a certain embodiment, specifically:

[0103] Analyze the water supply in normal and flood years, calculate the output monthly, and the abandoned water output values ​​N1 abandoned, N2 abandoned, N3 abandoned, N4 abandoned, and N5 abandoned. Calculate the annual utilization hours h1, h2, h3, h4, and h5 under these five types of abandoned water respectively;

[0104] Draw a curve of the relationship between abandoned water output and the average continuous output time over many years, calculate the economic annual utilization hours heconomic, and find the point S (Nabandoned, heconomic) in the curve. The abandoned water output Nabandoned is the repeated capacity;

[0105] The calculation formula for economic annual utilization hours h economic is:

[0106]

[0107] k 水 The cost of adding new generating capacity to a hydropower station, p 水 is the annual operating rate, i is the economic rate of return on repeated capacity, a is the capacity coefficient, and f is the output coefficient.

[0108] Step S22d, summing up the maximum working capacity of the hydropower station, the standby capacity of the hydropower station and the redundant capacity of the hydropower station to obtain the primary installed capacity of the hydropower station.

[0109] like Figure 4 As shown, according to one aspect of the present application, step S3 is further:

[0110] Step S31: construct a decision model based on heuristic rules based on Transformer, and construct the connection rules between the influencing factors of the installed capacity of each part of the hydropower station and its installed capacity;

[0111] Factors affecting the maximum working capacity of a hydropower station: reservoir water level, discharge flow, duration of water storage period and duration of water supply period;

[0112] Factors affecting the reserve capacity of a hydropower station: maximum load of the hydropower station;

[0113] Factors affecting the redundant capacity of hydropower stations: the power generation flow, the cost of the newly added power generation capacity of the hydropower station, the annual operating rate and the economic return rate of the redundant capacity;

[0114] Formulate heuristic rules to establish the relationship between the factors affecting the installed capacity of each part of the hydropower station and its installed capacity;

[0115] Maximum working capacity of hydropower station N 工 =f(Z, q, V, T 蓄 , T 供 )

[0116] Hydropower station reserve capacity N 备 =f(Nmax)

[0117] Hydropower station repeat capacity N 重 =f(q 电 , k 水 , p 水 , i)

[0118] Where Z is the reservoir water level, q is the discharge flow, V is the reservoir water storage capacity, T 蓄 is the duration of water storage period, T 供 is the duration of the water supply period, Nmax is the maximum load of the hydropower station, q 电 is the power generation reference flow, k 水 The cost of adding new generating capacity to a hydropower station, p 水 is the annual operating rate, and i is the economic rate of return on repeated capacity.

[0119] Step S32, taking the spatial unit impact factor of the hydropower station in each similar watershed unit as the original data set, taking the first 80% of the original data set as the training set, and the last 20% as the verification set;

[0120] Step S33, input the training set into the decision model, compare the obtained result with the primary installed capacity of the hydropower station and adjust the decision model parameter weights based on the comparison result;

[0121] Step S34, extracting the validation set and inputting it into the trained decision model, solving the model to obtain the simulated installed capacity of the hydropower station and comparing it with the installed capacity of the hydropower station, and determining the accuracy of the decision model based on the comparison result, optimizing the decision model when the accuracy of the decision model is less than a threshold, repeatedly iterating until the accuracy of the decision model is greater than the threshold, and obtaining the optimized decision model;

[0122] Step S35, extract the influencing factors of the hydropower station spatial unit and input them into the optimized decision model, solve the model to obtain the ultimate installed capacity of the hydropower station.

[0123] In this embodiment, a combination of advanced algorithms and heuristic rules is adopted to greatly improve the accuracy and decision-making efficiency of the model, and the relevant data of the hydropower station can be analyzed and processed quickly and accurately, providing a reliable decision-making basis for the calculation of installed capacity. It solves the problems of inaccurate decision-making and low efficiency of traditional models when faced with complex hydropower station data and actual operating conditions, and improves the practicality and reliability of the model.

[0124] The Transfer Learning method can effectively utilize existing knowledge, reduce training time and data requirements, and continuously improve the performance of the model through comparison and optimization with the validation set, so that it can better adapt to different data sets and actual application scenarios. It solves the problems of long model training time, large amount of data support and poor generalization ability in similar methods, and improves the adaptability and reliability of the model.

[0125] like Figure 5 As shown, according to one aspect of the present application, step S4 is further:

[0126] Step S41, constructing an optimal dispatching model for a hybrid hydropower station group, wherein the objective function is to maximize power generation, the decision variable is the downstream flow rate, and the constraints include: water balance of each reservoir, upper and lower water level constraints of each reservoir, discharge capacity of each reservoir discharge gate, downstream safety discharge, flood discharge gate operation constraints and non-negative constraints, and taking the installed capacity of the hydropower station as input, using the fish algorithm to solve the model and obtain the maximum power generation;

[0127] Taking the maximum power generation as the objective function, the optimal dispatching model of hybrid hydropower station is established:

[0128] =η×ρ×g×H×Q

[0129] Among them, E is the total power generation of the hydropower station cluster, η is the turbine efficiency, ρ is the density of water, g is the gravitational acceleration, H is the head, and Q is the flow rate;

[0130] Set the decision variables and hard constraints, specifically:

[0131] Decision variables: downstream flow rate Q;

[0132] The water balance equation of each reservoir is:

[0133] Vt end - Vt beginning = (Q in - Q out) t

[0134] Among them, Vtbegin and Vtend are the reservoir capacities at the beginning and end of the time period, Qin and Qout are the inflow and outflow flows of the time period, and t is the time of the time period;

[0135] The upper and lower water level constraints of each reservoir: Zmin <Z<Zmax

[0136] Wherein, Z is the average water level in front of the dam at time t of the reservoir, Zmin and Zmax are respectively the lowest and highest water levels allowed during the reservoir operation period, the highest water level is the flood control limit water level, and the lowest water level is the dead water level;

[0137] Discharge capacity constraint of each reservoir discharge gate: Qdischarge min < Qdischarge < Qdischarge max

[0138] Wherein, is the average discharge flow at time t of the reservoir, Qdischarge min and Qdischarge max are respectively the minimum and maximum discharge flows of the discharge gate;

[0139] Downstream safety discharge constraint: Qsafety < Qdischarge

[0140] Operation constraint of flood discharge gate: symmetric opening, power generation gate常开 during flood season, and the lowest opening water level of spillway gate;

[0141] Non - negativity constraint: Each physical quantity in the model is a non - negative real number.

[0142] Step S42: Collect the influencing factors of the hydropower station cost and conduct sensitivity calculation. Based on the sensitivity magnitudes of each influencing factor of the hydropower station cost, use the step - by - step element reduction analysis method to calculate the power generation and cost of the power station combination after each element reduction in turn;

[0143] Step S43: Draw the relationship curve between power generation and cost, and use the piece - wise linear fitting method to identify the critical point of the change of the relationship curve as the most economic point, and the corresponding hydropower station combination is the optimal installed capacity plan of the hydropower station.

[0144] Taking the cost as the abscissa and the power generation as the ordinate, plot the power generation and the corresponding cost after each element reduction in the coordinate system, and connect the points to obtain the power generation - cost relationship curve;

[0145] Analyze the change trend of the power generation - cost relationship curve, find the point where the slope of the curve changes significantly, use the piece - wise linear fitting method to divide the power generation - cost relationship curve into several segments, fit each segment with a straight line, calculate the slope of each straight line, and by comparing the slopes of adjacent two straight lines, the position where the slope changes greatly is the critical point.

[0146] According to one aspect of the present application, the step S41 is further as follows:

[0147] Step S41a: Set the population size, the maximum number of iterations, and the maximum moving step length of the remora fish, initialize the individual positions, randomly generate candidate solutions, and select the current candidate solution as the optimal solution;

[0148] Step S41b, adjusting the candidate solution position based on the host position, inputting each individual into the hybrid hydropower station group optimization scheduling model, and calculating the corresponding total power generation, which is the fitness value of the individual;

[0149] Step S41c: for each individual, based on the fitness value of the individual, select a better individual as the follow-up object, update the current optimal solution and the candidate solution position, and make a judgment. When the fitness of the candidate solution is better, update the global optimal solution;

[0150] Step S41d, repeatedly iterate until the maximum number of iterations is reached, obtain the scheduling plan corresponding to the individual with the maximum fitness, and calculate the maximum power generation.

[0151] The remora algorithm has a strong global search capability and can quickly find the optimal solution in a complex solution space, which improves the efficiency and accuracy of the solution and provides effective algorithm support for the hydropower station to achieve maximum power generation. The algorithm simulates the parasitic and foraging behavior of the remora. First, the individual position is initialized, candidate solutions are randomly generated, and the current candidate solution is selected as the optimal solution. Then the position of the candidate solution is adjusted based on the host position. In this process, the candidate solution will move and adjust according to the host's position information and certain rules, and the fitness value after each adjustment is calculated. The fitness value reflects the power generation corresponding to the candidate solution. Then the current optimal solution and the candidate solution position are updated and judged. When the fitness of the candidate solution is better, the global optimal solution is updated. Through repeated iterations until the maximum number of iterations is reached, the optimal solution finally obtained is the maximum power generation.

[0152] In a certain embodiment, specifically:

[0153] It is assumed that there are 50 fish in the search space at the beginning, and the position of each fish is set as the upper and lower limits of the outbound flow. Assuming that the outbound flow range of each fish is the upper and lower limits of the outbound flow, each fish randomly initializes its outbound flow in the search space and selects the current position of the fish as the optimal solution;

[0154] The fish will gradually approach the global optimal solution, that is, the outflow flow with the maximum power generation, so that the candidate solution approaches the current global optimal solution;

[0155] According to the position of the host fish and the positions of other random fish, the positions of the candidate solutions are adjusted using the aggregation factor and the dispersion factor, and the fitness value of the updated solution is calculated, specifically:

[0156] Calculate the hydraulic head and power generation given the outflow flow;

[0157] When the fitness of the candidate solution is better, the global optimal solution is updated.

[0158] According to one aspect of the present application, step S42 is further:

[0159] Step S42a, collecting data on factors affecting the cost of a hydropower station, and constructing a mixed hydropower station cost function, wherein the factors affecting the cost of a hydropower station include: upstream and downstream head difference, water resource stability, geological conditions of the hydropower station, transmission line model, substation equipment model, concrete pouring method, generator set model, and turbine set model;

[0160] Step S42b, using the local perturbation method to perform sensitivity analysis on the factors affecting the cost of the hydropower station, obtain the sensitivity of the factors affecting the cost of each hydropower station, and generate the cost of each hydropower station as an initial solution based on the power generation of the hydropower station;

[0161] Step S42c, select the solution space and gradually reduce the dimensions of the hydropower station cost influencing factors in order from small to large according to the sensitivity of the hydropower station cost influencing factors to obtain the power station combination after each reduction, calculate the power generation and cost of the power station combination after the reduction, and iterate repeatedly until the maximum number of iterations.

[0162] According to the different power generation of the hydropower station, the cost of each hydropower station is generated, the initial solution is generated, and the solution space is selected. The key to the step-by-step reduction method is to gradually reduce the dimension from the high-dimensional solution space and simplify the problem;

[0163] Gradually reduce the dimension of the objective function variables. At each step, selectively reduce the dimension of one or more decision variables, selectively find some variables that have less impact on the objective function, and reduce the amount of calculation by eliminating minor variables, making the problem-solving process more efficient. After each reduction step, use the optimization algorithm to solve the simplified problem. In each round of optimization, the goal of the solution is to obtain the local optimal solution for the current simplified problem. After each dimensionality reduction, check whether the current optimization solution still meets the constraints and ensure the quality of the solution.

[0164] The algorithm terminates when the maximum number of iterations is reached. If the quality of the solution does not improve significantly in a certain reduction step, or the constraints cannot be met, the backtracking mechanism can be used to restore to the previous solution and continue to adjust the strategy.

[0165] According to one aspect of the present application, step S43 is further:

[0166] Step S43a, with the construction cost as the horizontal coordinate and the power generation as the vertical coordinate, the power generation after each reduction and the corresponding construction cost are plotted in the coordinate system, and each point is connected to obtain a power generation-construction cost relationship curve;

[0167] Step S43b, using a piecewise linear fitting method to divide the power generation-cost relationship curve into several segments, each segment is fitted with a straight line, and the slope of each straight line is calculated;

[0168] Step S43c, calculate the change of the slope of the two straight lines before and after each data point in turn, take the data point corresponding to the maximum change as the critical point, and the hydropower station combination corresponding to the critical point is the optimal installation plan of the hydropower station.

[0169] In dealing with the relationship between the power generation and cost of a hydropower station and determining the optimal economic scale, the calculation process of the piecewise linear fitting method is as follows: first, the power generation and cost data of hydropower stations of different sizes are collected, and a scatter plot is drawn after preprocessing; then, the data is segmented assuming that there is a critical point, and a linear regression fitting line is used for each segment. By traversing the data points, the critical point is determined with the minimum total error as the criterion; in this process, the error of each segment of the fitting line is calculated to evaluate the fitting effect.

[0170] The principle is based on the assumption that the relationship between power generation and cost presents different linear trends at different stages, and the segmented fitting is used to more accurately characterize this complex relationship. The advantages of this method are significant. It can effectively handle the nonlinear relationship that may exist in the data. Compared with a single linear fitting, it can be more in line with the actual situation, and solves the problem that traditional linear fitting cannot accurately describe the complex relationship between power generation and cost, and avoids the deviation caused by simple linear assumptions. The advantage is that by determining the most economical point, a scientific basis is provided for the planning and construction of hydropower stations, helping decision makers to control the cost to the greatest extent while meeting the power generation demand, achieve optimal allocation of resources, and improve the economic benefits and feasibility of hydropower station projects. In this embodiment, by constructing a comprehensive cost function and a scientific analysis method, the balance point between power generation and cost can be intuitively found, the optimal economic scale can be accurately determined, and an important basis for investment decisions of hydropower stations is provided. The problem that similar methods cannot accurately determine the optimal economic scale, resulting in poor investment returns, is solved, and the economic benefits of hydropower station construction are improved.

[0171] This innovative method for determining the installed capacity plan for a group of hybrid hydropower stations in complex river basins provides key support for the planning and development of the hydropower industry with a comprehensive and scientific process. From collecting multi-dimensional data in the study area and conducting cluster analysis, to accurately calculating the output of hydropower stations in each river basin unit, to building an optimization decision-making model to solve the installed capacity, and finally determining the optimal economic scale, each step is closely linked and progressive. This method effectively overcomes the shortcomings of traditional methods in data collection, planning rationality, model accuracy and economic scale determination by relying on comprehensive consideration of complex factors, the use of advanced algorithms and scientific analysis processes. With the continuous growth of clean energy demand, this method is expected to be applied in more hydropower station construction projects in complex river basins, helping the hydropower industry to improve energy utilization efficiency and achieve a win-win situation of economic and environmental benefits, making important contributions to global sustainable energy development and promoting technological innovation and practical progress in the field of hydropower.

[0172] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A method for determining the installation plan of a complex river basin hybrid hydropower station group, characterized in that: The steps include: Step S1, determine the number of hydropower stations required in the study area, divide several hydropower station units, compress and cluster them, and form several similar watershed units; Step S2, calculating the number of hydropower stations and primary installed capacity in each similar watershed unit; Step S3: using a transformer-based decision model, taking the hydropower station and the initial installed capacity as input, to obtain the ultimate installed capacity of the hydropower station; Step S4, input the ultimate installed capacity of the hydropower station into the pre-built hybrid hydropower station group optimization scheduling model, use the fish algorithm to solve and obtain the maximum power generation and use it as the upper bound, deduct the hydropower station cost influencing factors in turn and calculate the corresponding power generation, draw the power generation-cost relationship curve, identify the critical point of the curve as the most economical point, and determine the optimal installed capacity plan of the hydropower station.

2. A method for determining a complex basin hybrid hydropower station installation plan as claimed in claim 1, characterized in that: The step S1 is further as follows: Step S11, collecting data of the study area, including hydrological data of the study area and spatial unit impact factors of hydropower stations; Step S12: determining the number of hydropower stations required in the study area based on the data of the study area, and dividing the determined hydropower stations into a number of hydropower station units based on the hydropower station spatial unit influencing factors; Step S13: construct an autoencoder that can optimize the objective function, and use several hydropower station spatial unit influencing factors as input, compress and embed several hydropower station spatial unit influencing factors into a low-dimensional feature space, and then use the DBSCAN algorithm to cluster them to form several similar watershed units.

3. A method for determining a complex basin hybrid hydropower station installation plan as claimed in claim 2, characterized in that: The step S13 is further as follows: Step S13a, constructing an optimizable objective function autoencoder based on the division of the impact factors of the hydropower station spatial unit, including: an encoder and a decoder, setting the domain radius and the minimum number of points; Step S13b, using an encoder to add a feature clustering layer based on the hydropower station spatial unit impact factor at the output, and compressing the hydropower station spatial unit impact factor and embedding it into a low-dimensional feature space; Step S13c, for each data point in the low-dimensional feature space, calculate the number of points in the domain and determine whether it is greater than the minimum number of points, and take the point greater than the minimum number of points as the core point; Step S13d, starting from a core point, all points within the core point area are added to the same cluster. When a point within the area is also a core point, the cluster is expanded with the point as the starting point until it cannot be expanded further, and several cluster centers are obtained; Step S13e, taking several cluster centers as the objective function, optimizing the objective function through back propagation and the autoencoder model, and obtaining several optimized cluster centers to divide the study area into several similar watershed units.

4. The method for determining the installation plan of a complex-basin hybrid hydropower station group according to claim 1, characterized in that: The step S2 is further as follows: Step S21, reading and calculating the number of hydropower stations in each similar watershed unit based on the watershed area, river length, natural river runoff, river network density, water system shape, land use pattern, soil type, temperature and sunshine time of each similar watershed unit; Step S22, respectively calculate the maximum working capacity, spare capacity and redundant capacity of the hydropower station in each similar watershed unit, and obtain the primary installed capacity of each hydropower station, which is the sum of the maximum working capacity, spare capacity and redundant capacity of the hydropower station.

5. A method for determining a complex basin hybrid hydropower station installation plan as claimed in claim 4, characterized in that: The step S22 is further as follows: Step S22a, collect historical data of the hydropower station to obtain the average flow in the dry year, calculate the storage capacity between the dead water level and the normal water level, that is, the beneficial storage capacity of the hydropower station, use a trial algorithm to calculate the water storage period, water supply period and power generation reference flow of the hydropower station, calculate the monthly average output of the hydropower station, that is, the guaranteed output of the hydropower station, calculate the guaranteed electric energy of the hydropower station based on the guaranteed output of the hydropower station and the water supply period of the reservoir, and obtain the maximum working capacity of the hydropower station; Step S22b, calculating the load reserve capacity and the accident reserve capacity based on the maximum load of the system, the reserve capacity of the hydropower station is the sum of the load reserve capacity and the accident reserve capacity; Step S22c, based on the water inflow in normal water years and flood water years, the output of the hydropower station is calculated monthly, and the relationship curve between the abandoned water output and the multi-year average continuous output time is drawn, and the economic annual utilization hours are calculated. The economic annual utilization hours are used as coordinate values, and the abandoned water output value obtained is the repeated capacity. Step S22d, sum the maximum working capacity of the hydropower station, the standby capacity of the hydropower station and the redundant capacity of the hydropower station to obtain the primary installed capacity of the hydropower station.

6. A method for determining a complex basin hybrid hydropower station installation plan according to claim 1, characterized in that: The step S3 is further as follows: Step S31: construct a decision model based on heuristic rules based on Transformer, and construct the connection rules between the influencing factors of the installed capacity of each part of the hydropower station and its installed capacity; Step S32, taking the spatial unit impact factor of the hydropower station in each similar watershed unit as the original data set, taking the first 80% of the original data set as the training set, and the last 20% as the verification set; Step S33, input the training set into the decision model, compare the obtained result with the primary installed capacity of the hydropower station and adjust the decision model parameter weights based on the comparison result; Step S34, extracting the validation set and inputting it into the trained decision model, solving the model to obtain the simulated installed capacity of the hydropower station and comparing it with the installed capacity of the hydropower station, and determining the accuracy of the decision model based on the comparison result, optimizing the decision model when the accuracy of the decision model is less than a threshold, repeatedly iterating until the accuracy of the decision model is greater than the threshold, and obtaining the optimized decision model; Step S35, extract the influencing factors of the hydropower station spatial unit and input them into the optimized decision model, solve the model to obtain the ultimate installed capacity of the hydropower station.

7. The method for determining the installation plan of a complex-basin hybrid hydropower station group according to claim 1, characterized in that: The step S4 is further as follows: Step S41, constructing an optimal dispatching model for a hybrid hydropower station group, taking the installed capacity of the hydropower station as input, and using the fish algorithm to solve the model to obtain the maximum power generation; Step S42: collect the cost influencing factors of the hydropower station and perform sensitivity calculation, and use the stepwise reduction analysis method to calculate the power generation and cost of the power station combination after each reduction based on the sensitivity of the cost influencing factors of each hydropower station; Step S43, draw a relationship curve between power generation and construction cost, and use piecewise linear fitting method to identify the critical point of the relationship curve as the most economical point. The corresponding combination of hydropower stations is the optimal installation plan of the hydropower station.

8. A method for determining a complex basin hybrid hydropower station installation plan as claimed in claim 7, characterized in that: The step S41 is further as follows: Step S41a, setting the population size, the maximum number of iterations and the maximum moving step length of the remora, initializing individual positions, randomly generating candidate solutions, and selecting the current candidate solution as the optimal solution; Step S41b, adjusting the candidate solution position based on the host position, inputting each individual into the hybrid hydropower station group optimization scheduling model, and calculating the corresponding total power generation, which is the fitness value of the individual; Step S41c: for each individual, based on the fitness value of the individual, select a better individual as the follow-up object, update the current optimal solution and the candidate solution position, and make a judgment. When the fitness of the candidate solution is better, update the global optimal solution; Step S41d, repeatedly iterate until the maximum number of iterations is reached, obtain the scheduling plan corresponding to the individual with the maximum fitness, and calculate the maximum power generation.

9. A method for determining a complex basin hybrid hydropower station installation plan according to claim 7, characterized in that: The step S42 is further as follows: Step S42a, collecting data on factors affecting the cost of a hydropower station and constructing a hybrid hydropower station cost function; Step S42b, using the local perturbation method to perform sensitivity analysis on the factors affecting the cost of the hydropower station, obtain the sensitivity of the factors affecting the cost of each hydropower station, and generate the cost of each hydropower station as an initial solution based on the power generation of the hydropower station; Step S42c, select the solution space and gradually reduce the dimensions of the hydropower station cost influencing factors in order from small to large according to the sensitivity of the hydropower station cost influencing factors to obtain the power station combination after each reduction, calculate the power generation and cost of the power station combination after the reduction, and iterate repeatedly until the maximum number of iterations.

10. A method for determining a complex basin hybrid hydropower station installation plan according to claim 7, characterized in that: The step S43 is further as follows: Step S43a, with the construction cost as the horizontal coordinate and the power generation as the vertical coordinate, the power generation after each reduction and the corresponding construction cost are plotted in the coordinate system, and each point is connected to obtain a power generation-construction cost relationship curve; Step S43b, using a piecewise linear fitting method to divide the power generation-cost relationship curve into several segments, each segment is fitted with a straight line, and the slope of each straight line is calculated; Step S43c, calculate the change in the slope of the two straight lines before and after each data point in turn, take the data point corresponding to the maximum change as the critical point, and the hydropower station combination corresponding to the critical point is the optimal installation plan of the hydropower station.

Images