Improved multi-target cuckoo algorithm (IMOCS)-based economic method for water-light complementary scheduling system

Abstract

The invention discloses a multi-target cuckoo algorithm (IMOCS)-based economical method for a water-light complementary dispatching system, and belongs to the field of water-light multi-energy complementary power generation technologies (hydroelectric energy optimization). The method specifically comprises the steps that a constraint condition set serves as a solution space, an initial population is generated by initializing a multi-objective function, the multi-objective function is used for representing multiple optimization objectives of the water-light complementation system, the constraint condition set is used for representing multiple constraints of the water-light complementation system, and individuals in the population are used for representing the upstream water level of each reservoir; based on a multi-objective function, calculating individual fitness in the initial population and obtaining a non-dominated sorting result of the initial population; based on a non-dominated sorting result of the initial population, screening out non-inferior solutions from the initial population and putting the non-inferior solutions into an archiving set; and continuously iteratively updating the population based on the individual fitness in the population to be updated until an end condition is met, so that an optimal solution set can be quickly determined as a scheduling reference scheme of the water-optical complementary system, and multi-target optimization scheduling (also needing to be modified) is efficiently executed for the water-optical complementary system.

Description

[Technical field] The present application belongs to the field of hydro-photovoltaic multi-energy complementary power generation technology (hydropower energy optimization). The method specifically relates to an economic method for a hydro-photovoltaic complementary scheduling system based on an improved multi-objective cuckoo algorithm.

Technical Background

[0002] In order to achieve the optimal utilization of renewable resources, a multi-objective function is set for optimal scheduling, with the criteria of providing the power grid with the largest possible total power generation, the highest possible economic benefits and improving energy utilization. In actual scheduling, cascade scheduling must also consider the optimal management of reservoirs, and meet the needs of flood control, irrigation, shipping and other aspects through scientific scheduling. For this type of complex, high-dimensional multi-objective optimization problem of water-photovoltaic complementary systems, it is necessary to adopt an optimization algorithm with good convergence and not easy to fall into local optimality for solution. Based on the above background, the present invention adopts an economic method for water-photovoltaic complementary scheduling system based on an improved multi-objective cuckoo algorithm, comprehensively considers the maximization of economic benefits and the maximization of renewable energy utilization, and obtains the optimal water-photovoltaic complementary scheduling scheme. [Summary of the invention] In order to solve the above problems, the present invention proposes an economic method for a water-photovoltaic hybrid scheduling system based on an improved multi-objective cuckoo algorithm (IMOCS). Comprehensively considering minimizing the unit energy cost of the water-photovoltaic hybrid system and maximizing the utilization rate of renewable energy, three multi-objective functions are formulated, and the influence of factors such as the optimal management of cascade reservoirs, flood control, irrigation and ecology are considered at the same time. By proposing a non-dominated sorting algorithm (IMOCS-NSGA-Ⅲ) based on an improved multi-objective cuckoo algorithm, the convergence of the algorithm is improved, the time cost of calculation is greatly reduced, and the Pareto optimal frontier is obtained. The final optimal solution is selected from the obtained optimal solutions by combining field tests and expert experience.

[0004] In order to achieve the above technical objectives, the present invention is implemented by the following technical solutions:

[0005] Step 1: Establish a water-photovoltaic complementary system model using constraints and objective functions

[0006] A model for optimizing the operation of hydro-photovoltaic complementary scheduling is established, with the benefits of hydro-photovoltaic complementary power generation as the main focus, taking into account the water use for ecological environment, and providing the grid with the largest possible reliable output, minimizing unit energy cost and maximizing renewable energy utilization as the criteria. The following three objectives are selected, and the specific objective function is:

[0007] 1. The unit energy cost of water-solar complementary power is an important indicator for evaluating the economic feasibility of water-solar complementary power system, usually expressed as the average cost per kilowatt-hour of electricity. The steps for calculating the unit energy cost are as follows:

[0008] 1) Determine the total cost

[0009] The total cost includes initial investment cost, operation and maintenance cost, depreciation cost, and financing cost;

[0010] Initial investment cost: the total cost of system construction and equipment purchase, including PV panels, inverters, hydropower equipment, installation costs, etc.

[0011] Operation and maintenance costs: expenses for regular maintenance, troubleshooting, and daily management during system operation;

[0012] Depreciation cost: The depreciation cost of equipment, usually allocated based on the useful life of the equipment;

[0013] Financing costs: If financing is through loans or other means, interest expenses must also be considered.

[0014] 2). Forecasting power generation

[0015] Estimate the total electricity generated by the system over its lifetime (usually in kWh). Annual electricity generation can be estimated based on historical weather data, geographic location, and the efficiency of photovoltaic and hydroelectric generation technologies, among other factors.

[0016] 3). Calculate unit energy cost

[0017] The unit energy cost is calculated as:

[0018] Unit energy cost = total cost / total electricity generation

[0019] The total cost includes the sum of the present values ​​of all initial investments, operating and maintenance costs, and other related expenses, while the total power generation is the total power generation of the system during its life cycle.

[0020] 4) Consider the discount rate

[0021] In long-term investments, future cash flows need to be discounted, especially in projects with long life cycles. The discount rate can be used to calculate the present value of future operating and maintenance costs, and then the unit energy cost can be calculated.

[0022] Objective function 1: Minimize unit energy cost:

[0023]

[0024] Among them, c n is the unit energy cost of the nth-level hydropower station;

[0025] c g is the unit energy cost of the PV power station.

[0026] 2. The utilization rate of renewable energy in the water-photovoltaic complementary system is maximized, including maximizing the power generation of photovoltaic power stations and minimizing the amount of water abandoned by cascade hydropower stations.

[0027] Objective function 2:

[0028]

[0029] Objective function 3:

[0030]

[0031] N(n, t) = k n ·Q(n, t)·H(n, t)·g

[0032] Among them, f 2 is the total power generation of the photovoltaic power station;

[0033] f 3 is the total amount of water abandoned by cascade hydropower stations;

[0034] p t is the output of the photovoltaic power station at time t;

[0035] N_max(n) is the channel output limit of the nth-level hydropower station;

[0036] N(n, t) is the output of the nth hydropower station in the tth time period;

[0037] H(n, t) is the average generating head of the nth hydropower station in the tth time period;

[0038] k n is the output coefficient of the nth-level hydropower station.

[0039] The constraints of the water-photovoltaic complementary system are:

[0040] 1. Water balance constraints

[0041]

[0042] Where: is the initial reservoir water storage capacity of the nth-level power station in the t+1 period;

[0043] V n，t is the initial reservoir storage capacity of the nth power station in period t;

[0044] q n，t is the inflow flow of the nth power station during period t;

[0045] Q n，t is the power generation flow of the nth power station during period t;

[0046] S n，t is the water discharge of the nth power station during period t;

[0047] O n，t Other water consumption for the nth power station during period t;

[0048] Δt is the length of the calculation period.

[0049] 2. Reservoir water level constraints

[0050]

[0051] in: Z n，t is the lower limit of the water level at the beginning of period t at the nth power station;

[0052] is the upper limit of the water level at the beginning of period t at the nth power station;

[0053] Z n，t is the reservoir water level of the nth power station at the beginning of period t.

[0054] 3. Reservoir discharge flow constraints

[0055]

[0056] in: is the minimum downstream flow rate that should be guaranteed by the n-th power station during period t;

[0057] is the discharge flow of the nth power station during period t;

[0058] It is the maximum allowable discharge flow of the n-th power station in period t.

[0059] 4. Power station output constraints

[0060]

[0061] Where: N n，minis the minimum permissible output of the nth-level power station (depending on the type and characteristics of the turbine);

[0062] N n，max is the installed capacity of the nth power station.

[0063] The meanings of other symbols are the same as before.

[0064] 5. Channel Constraints

[0065] Do not exceed the maximum transmission capacity of the power station output channel.

[0066] 6. Non-negative constraints

[0067] All the above variables are non-negative (≥0).

[0068] Step 2: Determine the optimization method based on the objective function

[0069] The optimization method is an improved multi-objective cuckoo algorithm (IMOCS). The algorithm combines the non-dominated sorting algorithm NSGA-III to improve the convergence of the algorithm, combines the multi-objective function to calculate the fitness value, and uses the concept of Pareto advantage to find solutions to multi-objective problems. The optimization process is:

[0070] Step 1: Initialize the algorithm parameters, including the dimension of the solution, the maximum number of iterations, the population size, the size of the external archive set, and the upper and lower limits of the parameters to be optimized in the decision variables;

[0071] Step 2: Randomly generate the initial positions of individuals in the population according to the upper and lower limits of the decision variables and start the first iteration;

[0072] Step 3: Calculate and evaluate the fitness value of the current bird's nest;

[0073] Step 4: Generate a new cuckoo individual position through Lévy flight and evaluate the fitness value of the new solution. If the fitness of the new solution is better than the original position, update the cuckoo position.

[0074] Levy flight is a random walk with the characteristics of long jumps and short steps. Its mathematical model is:

[0075] x i (t+1)=x i (t)+α·Levy(λ)

[0076] Where: x i (t) is the position of the i-th cuckoo in the t-th generation; α is the step size scaling factor; Levy(λ) is the random search path, and the relationship between Levy(λ) and the flight time t follows the Levy distribution, that is:

[0077] Levy(λ)~u=t-λ , 1≤λ≤3

[0078] The Levy distribution can be generated by the following formula:

[0079] Levy(λ)=|v|1 / βu

[0080] Among them, u~N(0,σ u2 ); v~N(0,σ v2 );σ u =(Γ((1+β) / 2)·β·2(β-1) / 2Γ(1+β)·sin(πβ / 2))1 / β

[0081] Step 5: Use P a The positions of some bird nests are randomly replaced with probability to increase the diversity of the population and retain the best few bird nests at present.

[0082] Step 6: Find the current Pareto optimal solution set through non-dominated sorting.

[0083] The NSGA-III algorithm can control the overall distribution of the population through evenly distributed reference points, including multi-level hierarchical sorting, reference point selection, adaptive normalization and association operations with niche retention to maintain diversity and find high-quality solutions on the Pareto frontier.

[0084] Multi-level hierarchical sorting: NSGA-III introduces multi-level hierarchical sorting, which divides the population into multiple hierarchical levels, each level containing solutions of different densities;

[0085] Reference point selection: NSGA-III selects a set of reference points evenly distributed in the target space to measure the superiority of the solution in the target space. These reference points are equivalent to marks in the solution space to guide the subsequent solution search algorithm;

[0086] Adaptive normalization: normalize the objective function value of each individual to adapt to the dimension and value range of different objective functions;

[0087] Association operation and niche preservation: The solution is associated with the reference point through the association operation, and the niche preservation operation is used to maintain the diversity of the solution;

[0088] Step7: Repeat the above steps until the maximum number of iterations is reached or the convergence condition is met.

[0089] Step 3: Establish a comprehensive status evaluation system to select the final optimal solution

[0090] Based on the degradation index and expert experience, a comprehensive state evaluation system is established to screen the Pareto frontier and determine the final optimal solution. When describing the evaluation system, the proportion of each indicator in the multi-objective function is considered.

[0091] Step 4: Design the Environmental Impact Assessment Module

[0092] Environmental impact assessment can calculate the impact of different scheduling schemes on the environment in real time during the scheduling optimization process. The construction period mainly focuses on land occupation, soil erosion, vegetation damage, etc.; the operation period mainly focuses on light blocking, water temperature changes, water quality impact, etc.; the decommissioning period mainly focuses on photovoltaic panel recovery, land restoration, etc.

[0093] The present invention aims at establishing a water-photovoltaic complementary scheduling model, establishes an objective function model that considers minimizing unit energy cost and maximizing renewable energy utilization, and adopts an improved multi-objective cuckoo optimization algorithm (IMOCS) for optimization, which can find the most economical solution for the water-photovoltaic complementary system under different situations.

Brief Description of the Drawings

[0094] Figure 1 A model for scheduling cascade hydropower stations and photovoltaic power stations established for the present invention;

[0095] Figure 2 The invention provides an economic method for a water-photovoltaic complementary scheduling system based on an improved multi-objective cuckoo algorithm (IMOCS) provided in an embodiment of the present application;

[0096] Figure 3 It is the distribution of 15 structured reference points of the three-objective problem when the average value of each dimension variable in the hyperplane is 4;

[0097] Figure 4 It is an optimization process based on the improved multi-objective cuckoo algorithm (IMOCS). [Specific implementation method]

[0098] The specific implementation methods of the present invention are described in more detail below in conjunction with the above-mentioned drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the protection scope of the present invention.

[0099] To illustrate the effect of the present invention, the following takes the medium- and long-term scheduling of a water-solar complementary system as the implementation object of the present invention, and describes the application principle of the present invention in detail in conjunction with the accompanying drawings:

[0100] The present invention provides an economic method for water-solar hybrid scheduling system based on improved multi-objective cuckoo algorithm (IMOCS), see Figure 1 , including the following steps:

[0101] Step 1: Establish a model for scheduling cascade hydropower stations and photovoltaic power stations

[0102] Cascade hydropower stations and photovoltaic power stations such as Figure 1 As shown in the figure, a scheduling model for the water-photovoltaic complementary system is established considering multi-objective functions and constraints.

[0103] Step 2: Determine the objective function and constraints

[0104] The terms "first" and "second" etc. regarding multiple objective functions in the specification and claims of this article are used to distinguish different objects. For example, the 1st, 2nd, and 3rd objective functions are used to distinguish different objective functions, rather than to describe the specific order of the objective functions.

[0105] 1. Objective Function

[0106] Objective function 1: Minimize unit energy cost:

[0107]

[0108] Among them, c n is the unit energy cost of the nth-level hydropower station;

[0109] c g is the unit energy cost of the PV power station.

[0110] The utilization rate of renewable energy in the hydro-photovoltaic complementary system is maximized, including maximizing the power generation of photovoltaic power stations and minimizing the amount of water abandoned by cascade hydropower stations.

[0111] Objective function 2:

[0112]

[0113] Objective function 3:

[0114]

[0115] Among them, f 2 is the total power generation of the photovoltaic power station;

[0116] f 3 is the total amount of water abandoned by cascade hydropower stations;

[0117] p t is the output of the photovoltaic power station at time t;

[0118] N_max(n) is the channel output limit of the nth-level hydropower station;

[0119] N(n, t) is the output of the nth hydropower station in the tth time period;

[0120] H(n, t) is the average generating head of the nth hydropower station in the tth time period;

[0121] k n is the output coefficient of the nth-level hydropower station.

[0122] 2. The constraints of the water-photovoltaic complementary system are:

[0123] 1). Water balance constraints

[0124]

[0125] Where: is the initial reservoir water storage capacity of the nth-level power station in the t+1 period;

[0126] V n，t is the initial reservoir storage capacity of the nth power station in period t;

[0127] q n，t is the inflow flow of the nth power station during period t;

[0128] Q n，t is the power generation flow of the nth power station during period t;

[0129] S n，t is the water discharge of the nth power station during period t;

[0130] O n，t Other water consumption for the nth power station during period t;

[0131] Δt is the length of the calculation period.

[0132] 2). Reservoir water level constraints

[0133]

[0134] Where: Z n，t is the lower limit of the water level at the beginning of period t at the nth power station;

[0135] is the upper limit of the water level at the beginning of period t at the nth power station;

[0136] Z n，t is the reservoir water level of the nth power station at the beginning of period t.

[0137] 3). Reservoir discharge flow constraints

[0138]

[0139] in: is the minimum downstream flow rate that should be guaranteed by the n-th power station during period t;

[0140] is the discharge flow of the nth power station during period t;

[0141] It is the maximum allowable discharge flow of the n-th power station in period t.

[0142] 4). Power station output constraints

[0143]

[0144] Where: N n，min is the minimum permissible output of the nth-level power station (depending on the type and characteristics of the turbine);

[0145] N n，max is the installed capacity of the nth power station.

[0146] The meanings of other symbols are the same as before.

[0147] 5). Channel constraints

[0148] Do not exceed the maximum transmission capacity of the power station output channel.

[0149]

[0150] 6). Non-negative condition constraints

[0151] All the above variables are non-negative (≥0).

[0152] Step 3: Optimization process based on improved multi-objective cuckoo algorithm (IMOCS)

[0153] Figure 2 The present invention provides an economic method for a water-photovoltaic hybrid scheduling system based on an improved multi-objective cuckoo algorithm (IMOCS). As shown in the figure, the method includes the following steps 1 to 7.

[0154] Step 1: Initialize the algorithm parameters, including the dimension of the solution, the maximum number of iterations, the population size, the size of the external archive set, and the value range of the parameters to be optimized in the decision variables;

[0155] Step 2: Randomly generate the initial positions of individuals in the population according to the upper and lower limits of the decision variables and start the first iteration;

[0156] Step 3: Calculate and evaluate the fitness value of the current bird's nest;

[0157] Step 4: Generate a new cuckoo individual position through Levy flight and evaluate the fitness value of the new solution. If the fitness of the new solution is better than the original position, update the individual position;

[0158] Step 5: Use P a Replace some bird nests randomly with probability, and keep the best ones.

[0159] Step 6: Find the current Pareto optimal solution set through non-dominated sorting;

[0160] Step 7: Repeat the above steps until the maximum number of iterations is reached or the convergence condition is met.

[0161] Figure 3 It is the distribution of 15 structured reference points of the three-objective problem when the average value of each dimension variable in the hyperplane is 4.

[0162] Figure 4 It is an optimization process based on the improved multi-objective cuckoo algorithm (IMOCS).

[0163] Step 4: Environmental Impact Assessment Module

[0164] In the example, the environmental impact assessment designed can calculate the impact of different scheduling schemes on the environment in real time during the scheduling optimization process. During the construction period, the main focus is on land occupation, soil erosion, vegetation destruction, etc.; during the operation period, the main focus is on light blocking, water temperature changes, water quality impact, etc.; during the decommissioning period, the main focus is on photovoltaic panel recovery, land restoration, etc.

[0165] It is easy for those skilled in the art to understand that the present invention is not limited to the above-mentioned embodiments. The above-mentioned embodiments and descriptions are only for illustrating the principles and advantages of the present invention. Various modifications and changes to the present invention without departing from the spirit and scope of the present invention are all included in the protection scope of the present invention.

Claims

1. An economic method for water-solar hybrid scheduling system based on improved multi-objective cuckoo algorithm (IMOCS), characterized by: The following steps are involved: Establish a model for optimal operation of water-solar complementary scheduling, i.e., a scheduling model for cascade hydropower stations and photovoltaic power stations; The peak load regulation model is constrained based on the constraint conditions of water-solar complementarity, and the constraint condition set includes: water balance constraint, reservoir water level constraint, reservoir discharge flow constraint, power station output constraint, channel constraint and non-negative condition constraint; The constraints of hydropower generation and photovoltaic power generation are optimized by using the IMOCS algorithm, and the initial positions (solutions) of individuals in the population are randomly generated. The individuals in the population represent the initial water levels of reservoirs at all levels. The fitness value of each individual in the population is determined based on the multi-objective function. The non-dominated sorting algorithm combined with the NSGA-III algorithm is used to determine the advantages and disadvantages of the solutions in the solution set, and individuals with greater crowding are selected based on the crowding distance, and relatively densely distributed solutions in the population are eliminated. New individual positions are generated based on Levy flight, and some individual positions are adjusted using chaotic mapping and elite strategy to increase the randomness of the algorithm. The new generation population and the population after the individual positions are updated are merged to obtain the merged population as the population to be updated in the next round of iteration, and the fitness value of the new solution is evaluated, and the low-quality solution is replaced by a new random solution. It is judged whether the end condition is met. If so, the solution set is output as the optimal solution set, otherwise the population is continuously updated iteratively, and the individuals in the optimal solution set are used as the reservoir scheduling reference scheme of the water-photovoltaic complementary system. Establishing a multi-objective function to characterize multiple optimization objectives of the water-photovoltaic complementary system, the multi-objective function includes: considering minimizing unit energy cost and maximizing renewable energy utilization rate, the first objective function takes minimizing unit energy cost as the goal, the second objective function takes maximizing the total power generation of photovoltaic power stations as the goal, and the third objective function takes minimizing the total amount of abandoned water in cascade hydropower stations as the goal; Solving the water-solar complementary model and obtaining a scheduling method; The termination condition is met when the maximum number of iterations is reached.

2. The economic method of water-photovoltaic complementary scheduling system based on IMOCS according to claim 1 is characterized in that: Taking into account the economic benefits and renewable energy utilization rate, a multi-objective optimization objective function for water-solar complementarity is established, including: Objective function 1: Minimize unit energy cost Calculating the unit energy cost includes considering the total cost, predicting the power generation, calculating the unit energy cost, and considering the discount rate. Then the unit energy cost is selected as the objective function 1: Min: Among them, c n is the unit energy cost of the nth-level hydropower station; c g is the unit energy cost of the PV power station. Objective functions 2 and 3: Maximizing the utilization of renewable energy The photovoltaic output power generation and the water abandonment of cascade hydropower stations are selected as objective function 2 and objective function 3 respectively: Max: Min: Q q (n,t)=(N_max(n)-N(n,t)) / (H(n,t)*k n ) N(n,t)=k n ·Q(n,t)·H(n,t)·g Among them, f2 is the total power generation of the photovoltaic power station; f3 is the total amount of abandoned water from cascade hydropower stations; p t is the output of the photovoltaic power station at time t; N_max(n) is the channel output limit of the nth-level hydropower station; N(n,t) is the output of the nth hydropower station in the tth time period; H(n,t) is the average generating head of the nth hydropower station in the tth time period; k n is the output coefficient of the nth-level hydropower station.

3. The economic method of water-photovoltaic complementary scheduling system based on IMOCS according to claim 2 is characterized in that: Combined with the non-dominated sorting algorithm of NSGA-Ⅲ algorithm, the solution set is sorted, and the non-dominated level and crowding distance are used to perform non-dominated sorting to determine the quality of the solutions in the solution set. Including: Using the concept of non-dominated sorting, individuals are divided into different levels. The target space is discretized and divided into several hyperplanes by introducing reference points. Each reference point represents a target combination. These reference points evenly cover the target space and are used to guide the algorithm to search for a better solution and ensure a uniform distribution of solutions. Select a predefined structured way to generate reference points and place the reference points in a normalized hyperplane as shown in the following formula: Among them, M is the number of optimization targets, H is the total number of reference points, and h is the optimization target score segment. The crowding degree of an individual is evaluated by calculating the distance between the individual and its neighboring reference points, and the crowding degree of an individual is calculated to reflect the density of each individual around the reference point. In the selection phase, non-dominated solutions are preferred. When selecting from non-dominated solutions based on crowding distance, individuals with larger distances are preferred to ensure that the selected individuals are evenly distributed in the target space.

4. The economic method of water-photovoltaic complementary scheduling system based on IMOCS according to claim 3 is characterized in that: The system adds an environmental impact assessment module that can calculate the impact of different scheduling schemes on the environment in real time during the scheduling optimization process. During the construction period, the main concerns are land occupation, soil erosion, vegetation destruction, etc.; during the operation period, the main concerns are light blocking, water temperature changes, water quality impacts, etc.; during the decommissioning period, the main concerns are photovoltaic panel recovery, land restoration, etc.

Images