Hydroelectric generating set combination optimization method based on genetic particle swarm hybrid algorithm

Abstract

The invention relates to a hydroelectric generating set combination optimization method based on a genetic particle swarm hybrid algorithm. The method comprises the following steps: obtaining a water consumption-energy consumption function of a hydroelectric generating set; constructing a hydroelectric generating set energy consumption optimization model based on the energy consumption function of each hydroelectric generating set in combination with an exterior point penalty mode; the hybrid algorithm adopts a nested structure, the outer layer adopts a genetic algorithm to optimize unit start and stop, and the inner layer adopts a particle swarm algorithm to optimize load distribution between started units. Compared with the prior art, the method has the advantages that the problems of unit start-stop and load distribution in unit combination are solved through the genetic particle swarm hybrid algorithm, the nested structure of the hybrid algorithm is simple in hierarchy and easy to implement, and minimization of operation energy consumption of the hydroelectric generating set is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of energy conservation of hydroelectric generating units, and particularly to a combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm. Background Art

[0002] The daily load optimal distribution of a hydropower station includes two core problems: unit commitment and load distribution among units. The purpose is to reasonably arrange the start-stop modes and output plans of each unit under the premise of meeting the system load demand and complex scheduling constraints. The combined optimization of hydropower station units can not only provide a decision-making basis for formulating short-term power generation scheduling plans for hydropower stations, but also guide hydropower stations to optimally execute the power generation tasks assigned by the power grid. However, in actual operation, for some large-scale hydropower stations, the hydropower operation presents typical characteristics such as high-dimensional, non-linear, discrete, and non-convex, which makes it very difficult to model and solve the daily load optimal distribution.

[0003] The unit commitment problem of a hydropower station simultaneously includes 0-1 integer variables representing the start-stop states of units and continuous variables representing the output of units. The genetic algorithm uses binary variables to optimize the objective function and has an advantage in optimizing discrete start-stop variables composed of 0-1. The particle swarm algorithm is a commonly used integer variable optimization algorithm. The structures of both algorithms are simple and easy to implement, and they have characteristics such as parallel search, memory ability, and fast convergence speed, and are widely used in the optimization field. Through example verification, it is shown that under the premise of meeting the load demand, this hybrid algorithm can solve the optimal start-stop unit combination and the load distribution among units to achieve the effect of reducing energy consumption. Summary of the Invention

[0004] The purpose of the present invention is to solve the two problems of unit start-stop and load distribution in the field of combined optimization of hydroelectric generating units through an intelligent optimization algorithm, so as to realize the combined optimization of hydroelectric generating units and achieve the purpose of saving energy consumption.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm, comprising the following steps:

[0007] S1. Obtain the water consumption - energy consumption function of the hydroelectric generating unit, and construct an optimization function based on an additional penalty function mechanism;

[0008] S2. Use the genetic algorithm as the outer algorithm to optimize the unit start-stop combination with discrete binary factors;

[0009] S3. Use the particle swarm algorithm as the inner algorithm to optimize the optimal load distribution among the started units according to the optimization functions of each unit.

[0010] Furthermore, the specific process of constructing the optimization function of the hydropower unit in step S1 is as follows:

[0011] S11. The water consumption - energy consumption function of the hydropower unit is:

[0012]

[0013] where P i,t is the energy consumption function of the i - th unit at time t, q i,t is the load of the i - th unit at time t, a, b, c are the coefficients of the energy consumption function, q i,min and q i,max are the minimum and maximum loads of the i - th unit.

[0014] S12. The additional penalty function is specifically

[0015]

[0016] where q t is the total demand load at time t, D is the total number of units, λ is the penalty coefficient, usually an infinite value;

[0017] Therefore, the optimization function of the hydropower unit in step S1 is specifically:

[0018] minP t =min[P i,t +Y]

[0019] where P t is the total water consumption at time t.

[0020] Furthermore, the specific process of using the genetic algorithm to optimize the start - stop combination of hydropower units in step S2 is as follows:

[0021] S21. Randomly generate N D - dimensional binary individuals, and each individual represents a set of start - stop combinations of hydropower units. For example, if D = 5 and an individual is

[01010] , it represents starting the second and fourth units;

[0022] S22. Pass the N binary individuals to the inner - layer particle swarm algorithm, and the particle swarm algorithm initializes the algorithm parameters according to different individuals;

[0023] S23. Receive the optimization results of the inner - layer particle swarm algorithm, and adjust the binary individuals through genetic algorithm processes such as selection, crossover, and mutation according to the results;

[0024] Furthermore, the specific process of using the particle swarm algorithm to optimize the load distribution among the starting units in step S3 is as follows:

[0025] S31. Initialize parameters such as particle positions, velocities, and optimization functions according to the binary individuals of the outer genetic algorithm;

[0026] S32. Obtain the optimal fitness value of each binary individual through the optimization mechanism of the particle swarm algorithm;

[0027] S33. Transmit the optimization result of the inner particle swarm algorithm to the outer genetic algorithm.

[0028] Compared with the prior art, the present invention has the following advantages:

[0029] Based on the problems of unit start-stop and load distribution in the combined optimization of hydropower units, the present invention combines the mechanism advantages of the genetic algorithm and the particle swarm algorithm in optimizing discrete variables and continuous variables, and constructs a nested structure with the particle swarm algorithm as the inner algorithm and the genetic algorithm as the outer algorithm. This structure is simple and easy to implement, and through example verification, it can achieve the purpose of optimizing the start-stop and load distribution of hydropower units, thereby achieving the goal of reducing energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic flow chart of the method of the present invention;

[0031] Figure 2 It is the load demand of the Three Gorges Hydropower Station on a certain day DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The present invention will be described in detail below with reference to the drawings and specific embodiments.

[0033] As Figure 1 shown, a method for combined optimization of hydropower units based on a genetic particle swarm hybrid algorithm includes the following steps:

[0034] S1. Obtain the water consumption - energy consumption function of the hydropower unit, and construct an optimization function based on the external penalty function mechanism;

[0035] S2. Use the genetic algorithm as the outer algorithm to optimize the unit start-stop combination with discrete binary factors;

[0036] S3. Use the particle swarm algorithm as the inner algorithm to optimize the optimal load distribution among the starting units according to the optimization functions of each unit.

[0037] This embodiment takes the actual power generation process of 26 units of the Three Gorges Hydropower Station at 24 time periods on a certain day as an example.

[0038] Step 1. Construct the energy consumption optimization function of the hydropower unit:

[0039] According to relevant data, the load - energy consumption function of the Three Gorges units is as shown in the following table:

[0040] Table 1 Load - Energy Consumption Parameters of Three Gorges Units under 95m Water Head

[0041]

[0042] The optimizable function is as follows:

[0043]

[0044] Among them, λ is assigned a positive infinity value.

[0045] Step 2: Initialize the initial parameters of the genetic - particle swarm hybrid algorithm as shown in the following table:

[0046] Table 2: Simulation Parameter Settings of the Improved Genetic Particle Swarm Hybrid Algorithm

[0047]

[0048] Step 3: Perform the combined optimization of hydropower units using the method of the present invention:

[0049] The load demand of the power station on a certain day is as Figure 1 shown. The unit load distribution after optimization by the present invention is shown in Table 3 - 1 and Table 3 - 2.

[0050] Table 3 - 1: Unit Load Distribution of Three Gorges on a Certain Day after Optimization by the Present Invention (Unit 1# - 13#)

[0051]

[0052] Table 3 - 2: Unit Load Distribution of Three Gorges on a Certain Day after Optimization by the Present Invention (Unit 14# - 26#)

[0053]

[0054]

[0055] It can be seen from Table 3 - 1 and Table 3 - 2 that the method of the present invention can determine the start - stop of units and optimize the load distribution among the started units on the premise of meeting the load demand. Taking the 5th moment as an example, Units 1#, 3#, 4#, 6#, 9#, 12#, 16#, 19#, and 25# are started respectively. The total water consumption is 477.683057×106 m3, which is lower than the total water consumption of 485.845624×106 m3 at this moment in the Three Gorges Hydropower Station. The water consumption saved is 8.162567×106 m3, and the energy - saving ratio is about 1.68%, thus verifying the effectiveness and superiority of the method of the present invention.

[0056] The above description of the embodiments is provided to enable those of ordinary skill in the art to understand and apply the present invention. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative efforts. Therefore, the present invention is not limited to the above embodiments, and all improvements and modifications made by those skilled in the art based on the disclosure of the present invention should fall within the protection scope of the present invention.

Claims

1. A combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm, characterized in that, it includes the following steps: S1. Obtain the water consumption - energy consumption function of the hydroelectric generating units, and construct an optimization function based on an additional penalty function mechanism; S2. Use the genetic algorithm as the outer - layer algorithm to optimize the start - stop combination of the generating units with discrete binary factors; S3. Use the particle swarm algorithm as the inner - layer algorithm to optimize the optimal load distribution among the starting generating units according to the optimization functions of each unit.

2. The combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm according to claim 1, characterized in that, the construction of the optimization function of the hydroelectric generating units in step S1 is specifically as follows: S11. The water consumption - energy consumption function of the hydroelectric generating units is: q i,min ≤q i,t ≤q i,max Among them, P i,t is the energy consumption function of the i-th unit at time t, q i,t is the load of the i-th unit at time t, a, b, and c are the coefficients of the energy consumption function, q i,min and q i,max are the minimum and maximum loads of the i-th unit; S12. The additional penalty function is specifically where q t is the total demand load at time t, D is the total number of units, and λ is the penalty coefficient; The optimization function of the hydroelectric generating units in step S1 is specifically: minP t = min[P i,t + Y] Among them, P t is the total water consumption at time t.

3. The combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm according to claim 1, characterized in that, the specific process of using the genetic algorithm to optimize the start - stop combination of the hydroelectric generating units in step S2 is: S21. Randomly generate N D - dimensional binary individuals, and each individual represents a set of start - stop combinations of the hydroelectric generating units; S22. Pass the N binary individuals to the inner - layer particle swarm algorithm, and the particle swarm algorithm initializes the algorithm parameters according to different individuals; S23. Receive the optimization results of the inner - layer particle swarm algorithm, and adjust the binary individuals through the genetic algorithm process according to the results.

4. The combined optimization method for hydroelectric generating units based on a genetic particle swarm hybrid algorithm according to claim 1, characterized in that, the specific process of using the particle swarm algorithm to optimize the load distribution among the starting generating units in step S3 is: S31. Initialize the particle positions, velocities, and optimization functions according to the binary individuals of the outer - layer genetic algorithm; S32. Obtain the optimal fitness value of each binary individual through the optimization mechanism of the particle swarm algorithm; S33. Pass the optimization results of the inner - layer particle swarm algorithm to the outer - layer genetic algorithm.

Images