CN116738841A - An algorithm and device for dynamic simulation of water energy in hydropower stations - Google Patents

Abstract

The invention discloses an algorithm and a device for dynamic simulation of water energy of a hydropower station, comprising the following steps: s1, acquiring natural working condition parameters of the operation of a hydroelectric generating set on line; s2, analyzing a comprehensive characteristic curve graph of the turbine model runner by utilizing data acquired by a neural network; s3, collecting relation curve data and output limit line data reflected in a comprehensive characteristic curve graph of the turbine model runner; s4, establishing an objective function of the unit operation condition efficiency C, and obtaining an optimizing solution; s5, describing a comprehensive characteristic curve graph of the turbine model runner, and analyzing by utilizing a neural network to collect data; s6, simulating the whole hydrologic process to obtain the average power generation amount for many years; and S7, obtaining an optimal solution by adopting a weighting algorithm according to the optimal solution and the average power generation quantity for many years. The invention combines the hydrological condition with the generating characteristic of the unit, simulates the real-time efficiency, improves the accuracy and avoids the situation that the generating characteristic of the unit is not consistent with the hydrological condition.

Description

An algorithm and device for dynamic simulation of water energy in hydropower stations

Technical field

The invention relates to the technical field of water conservancy and hydropower, and in particular to an algorithm for dynamic simulation of water energy in hydropower stations.

Background technique

A hydropower station is a comprehensive engineering facility that converts water energy into electrical energy. It generally includes a reservoir formed by water retaining and drainage buildings, a hydropower station water diversion system, a power plant building, and mechanical and electrical equipment. The high water level of the reservoir flows into the factory building through the water diversion system to drive the hydroelectric generator set to generate electrical energy, which is then input into the power grid through the step-up transformer, switch station and transmission line.

In the existing technology, the Chinese invention patent with the authorization announcement number CN103306886B discloses a dynamic simulation algorithm for hydropower station water energy. The patent adopts the following method steps: 1) Import the comprehensive characteristic curve diagram of the hydraulic turbine model runner into the database; Comprehensive characteristic curve diagram of the hydraulic turbine model runner, collect the characteristics reflected in the diagram Relationship curve data and output limit line data, import the collected data into the database; 2) Establish a neural network model, and through data training, enable it to reflect and describe the characteristics of the turbine/> relationship curve, the model meets sufficient accuracy requirements; 3) Import the long-term daily runoff series and hydraulic head data of the power station; 4) Import the turbine parameters (speed ni, diameter Di, installation elevation ▽); it is based on the given runner and hydrological process On the basis of this, the power generation process of the power station is mathematically simulated through an optimal intelligent algorithm model, and the annual power generation index of the power station is obtained. This provides a reliable basis for the modeling of the hydroelectric generator and the technical and economic analysis of the hydropower station.

However, in the existing technology, as the accuracy requirements for water energy calculations increase, the application of dynamic simulation methods in the field of water energy calculations will be gradually updated. If the effective combination of the operating conditions of the unit and the hydrological conditions cannot be comprehensively considered, it will The actual operation of hydropower stations causes a waste of water energy resources. The difficulty of using dynamic simulation methods to calculate electric energy lies in the diversity and complexity of the regulating capabilities and dispatching methods of hydropower stations. Some parameters of the calculation model need to be limited.

Therefore, it is necessary to propose an algorithm for dynamic simulation of water energy in hydropower stations to solve the above problems.

Contents of the invention

The present invention proposes an algorithm for dynamic simulation of water energy in hydropower stations to solve problems in the background technology. It combines artificial neural networks and genetic algorithms with water energy calculation models to perform calculations of water energy, unit selection and actual operation of hydropower stations. , establish a mathematical model of the factors affecting the water energy of the hydropower station, input the hydrological process parameters of the power station, simulate the basic situation of the power station during normal operation and power generation, and use historical records to calculate accurate power generation results through weighted algorithms to guide the selection of hydropower stations. The work can effectively combine hydrometeorological conditions with unit power generation characteristics, simulate real-time efficiency, improve accuracy, and avoid inconsistencies between unit characteristics and hydrological conditions.

The technical solution of the present invention is implemented as follows:

An algorithm for dynamic simulation of water energy in hydropower stations, including the following steps:

S1. Online collection of the natural working conditions parameters of the operation of the hydrogenerator unit: head loss H, unit reference flow rate Q, runner diameter D;

S2. Analyze the comprehensive characteristic curve of the hydraulic turbine model runner using the neural network to collect data and form a database;

S3. Collect the relationship curve data and output limit line data reflected in the comprehensive characteristic curve diagram of the hydraulic turbine model runner, and import the collected data into the database;

S4. Based on the turbine speed n and combined with the collected parameters in step S1, establish an objective function of unit operating condition efficiency C with unit flow rate A and unit speed B as parameters. According to the objective function and safety and stability constraints It is required to perform optimization with the database established by the neural network to obtain the optimal solution E ₁ = (A, B);

S5. Derive [Q _i (t), H _i (t)] at each time point in the long historical sequence of the unit, determine the diameter and rotation speed (D _j , n _j ) of the unit, and depict the comprehensive characteristic curve of the turbine model runner Figure, use neural network to collect data for analysis;

S6. By combining the invariants during the operation of a group of units with the hydrological process of the power station and reservoir dispatch control, and dynamically simulating the maximum power generation as the objective function, the hydrological process is simulated. The simulation process is, with power generation Dynamically simulate the objective function with the maximum quantity, use the neural network to collect data sets for analysis, input the hydrological processes such as water flow and upstream and downstream water levels of the hydropower station over the years into the parameters of the turbine model, and reselect (D _j ,n _j ) _m , The multi-year average power generation E ₂ =(A ₂ ,B ₂ ) is obtained;

S7. By optimizing the operating efficiency C of the unit, the requirements for safety and stability constraints and the established neural network database, the optimal solution E ₁ = (A, B) is obtained and the maximum power generation is used as the objective function. The multi-year average power generation E ₂ =(A ₂ ,B ₂ ) obtained by analyzing the neural network is finally used to solve E ₁ and E ₂ using a weighted algorithm to obtain the optimal solution E ^* .

To further optimize the technical solution, in step S4, the calculation formula of the unit flow rate A is:

The calculation formula of the unit speed B is:

To further optimize the technical solution, in step S4, the following model is used for the water energy calculation simulation of the hydropower station:

In the water energy calculation simulation model of hydropower stations,

Among them: H _1(t) is the water head corresponding to the upstream at time t, H _2(t) is the water head corresponding to the downstream at time t, ΔH is the head difference between the upstream and downstream, V ₀ is the initial water volume of the hydropower station, V(t) is the volume of water corresponding to the hydropower station at time t, β is the blade opening, β _max is the maximum blade opening, eta is the turbine efficiency, η _i is the generator efficiency, ∑ _t Q ⁱ is the diversion flow of all turbines, Q _max is the maximum overflow of the turbine, α is the guide vane opening, α _max is the maximum guide vane opening; D is the turbine efficiency set.

To further optimize the technical solution, in step S5, the output calculation model of the hydropower station is as follows:

Among them: Di is the diameter of the i-th turbine, H is the water head, α is the guide vane opening, β is the blade opening, Q _i is the unit flow rate of the i-th turbine, n _i is the unit speed of the i-th turbine, eta is the generator efficiency.

To further optimize the technical solution, in step S6, the calculation model of power generation is:

To further optimize the technical solution, in step S7, the calculation formula of the optimal solution E ^* is as follows:

Among them, E ^* =[(kA ₁ +rA ₂ ), (kB ₁ +rB ₂ )].

Adopting the above technical solution, the beneficial effects of the present invention are:

This invention combines the artificial neural network and genetic algorithm with the water energy calculation model to calculate the water energy of the hydropower station, unit selection and actual operation, establish a mathematical model of the influencing factors of the water energy of the hydropower station, input the hydrological process parameters of the power station, and simulate the normal operation of the power station. Basic conditions during power generation, combined with historical records, are used to calculate accurate power generation results through a weighted algorithm to guide the selection of hydropower stations. It can effectively combine hydrometeorological conditions with unit power generation characteristics, and simulate real-time efficiency. Improved accuracy to avoid inconsistencies between unit characteristics and hydrological conditions.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting any creative effort.

Figure 1 is a flow chart of an algorithm for dynamic simulation of water energy in hydropower stations according to the present invention.

Detailed ways

The technical solutions of the present invention will be described clearly and completely below with reference to specific embodiments. However, those skilled in the art should understand that the embodiments described below are only used to illustrate the present invention and should not be regarded as limiting the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

The dynamic simulation method is widely used and has many applications in control processes, mechanical simulations, and water intake. However, it is currently relatively rarely used in water energy calculations. The dynamic simulation method can constrain the operating conditions of the turbine. Effectively avoiding adverse operating conditions, as the accuracy requirements for water energy calculations increase, the application of dynamic simulation methods in the field of water energy calculations will gradually become popular.

The present invention proposes an algorithm for dynamic simulation of water energy in hydropower stations. As shown in Figure 1, it includes the following steps:

S1. Online collection of the natural working conditions parameters of the operation of the hydrogenerator unit: head loss H, unit reference flow rate Q, runner diameter D.

S2. Use the neural network (neural network to depict the unit operating curve) to collect data and analyze the comprehensive characteristic curve of the hydraulic turbine model runner to form a database.

S3. According to the comprehensive characteristic curve diagram of the hydraulic turbine model runner, collect the relationship curve data and output limit line data reflected in the diagram, and import the collected data into the database.

S4. Based on the turbine speed n and combined with the collected parameters in step S1, establish the unit flow rate and unit speed/> is the objective function of the parameter unit operating condition efficiency C. According to the objective function, the requirements of safety and stability constraints and the database established by the neural network, the optimization solution E ₁ = (A, B) is obtained.

When comparing water energy calculations and unit selection plans, the models and algorithms must meet consistency requirements, that is, each plan should be compared with other plans under its own optimal conditions, that is, "select the best among the best." Therefore, the mathematical model for water energy calculation and unit selection should contain three contents:

① Given the runoff and unit parameters, how to allocate the flow (output) of each unit so as to make full use of water energy.

② How to select unit parameters and flow distribution for a given runoff to maximize real-time power generation.

③ For long-sequence hydrological processes that take into account inter-annual changes, which model and corresponding invariant parameters should be selected to maximize the multi-year average power generation. From the basic principle of dynamic simulation method, the above three parts are inseparable and should be unified in a mathematical model. Experience shows that an effective tool for dealing with the above problems is the dynamic programming algorithm. Among them, the following model is used for water energy calculation and simulation of hydropower stations:

In the water energy calculation simulation model of hydropower stations,

Among them: H _1(t) is the water head corresponding to the upstream at time t, H _2(t) is the water head corresponding to the downstream at time t, ΔH is the head difference between the upstream and downstream, V ₀ is the initial water volume of the hydropower station, V(t) is the volume of water corresponding to the hydropower station at time t, β is the blade opening, β _max is the maximum blade opening, eta is the turbine efficiency, η _i is the generator efficiency, ∑ _t Q ⁱ is the diversion flow of all turbines, Q _max is the maximum overflow of the turbine, α is the guide vane opening, α _max is the maximum guide vane opening; D is the turbine efficiency set.

S5. Derive [Q _i (t), H _i (t)] at each time point in the long historical sequence of the unit, determine the diameter and rotation speed (D _j , n _j ) of the unit, and depict the comprehensive characteristic curve of the hydraulic turbine model runner. Use neural networks to collect data for analysis.

Among them, the output calculation model of hydropower stations is as follows:

Among them: Di is the diameter of the i-th turbine, H is the water head, α is the guide vane opening, β is the blade opening, Q _i is the unit flow rate of the i-th turbine, n _i is the unit speed of the i-th turbine, eta is the generator efficiency.

S6. Use the dynamic simulation method to combine the invariants during the operation of a group of units (specific rotation number, diameter, rotation speed, installation elevation, model runner comprehensive characteristics, etc.) with the hydrological process of the power station and reservoir dispatch control, using neural networks Data sets are collected for analysis, and hydrological processes such as water flow and upstream and downstream water levels of the hydropower station over the years are input into the model parameters. Dynamic simulation is performed with the maximum power generation as the objective function. All hydrological processes are simulated and re-selected (D _j , n _j ) _m , where (j=1,2,…,N,m=1,2,…,N), the multi-year average power generation E ₂ =(A ₂ ,B ₂ ) is obtained.

By using the dynamic simulation method, the invariants during the operation of a group of units are combined with the hydrological process of the power station and reservoir dispatch control. The dynamic simulation method is to put the invariants during the operation into a specific sequence of hydrological processes for testing. Finally, a "satisfactory solution" is selected from a comprehensive analysis of annual power generation, unit stability, cavitation performance, unit costs, and auxiliary facility costs. Finally, a neural network is used to collect data sets for analysis, and the water flow and flow rate of the hydropower station over the years are analyzed. Hydrological processes such as upstream and downstream water levels are input into the parameters of the turbine model, and the maximum power generation is used as the objective function to dynamically simulate the distribution of the flow (output) of each unit in order to fully utilize the water energy and re-select (D _j , n _j ) _m , we get the multi-year average power generation E ₂ =(A ₂ ,B ₂ );

In step S6, the calculation model of power generation is:

S7. Based on the optimal solution E ₁ = (A, B) and the multi-year average power generation E ₂ = (A ₂ , B ₂ ), a weighted algorithm is used to obtain the optimal solution E ^* . The calculation formula of the optimal solution E ^* is as follows:

Among them, E ^* =[(kA ₁ +rA ₂ ), (kB ₁ +rB ₂ )]. In order to improve the operating conditions of the hydropower station, it avoids reservoir water level fluctuations and frequent startup and shutdown operations, avoids no-load flow loss during shutdown, increases the operating range of the unit, and effectively utilizes water energy at low flow rates.

In view of the problem that traditional hydropower station water energy calculation often depends on experience, and the calculation results sometimes have large deviations, the present invention proposes a water energy dynamic simulation method, which takes the hydrological process, unit characteristics and dispatch control of the power station as a whole, and based on real-time engineering. Calculate the real-time power generation based on the efficiency of the operating conditions, and use the multi-year average power generation as the objective function to obtain the optimal unit parameters and thereby determine the unit operating parameters, which can achieve good results.

This invention combines artificial neural networks and genetic algorithms with water energy calculation models to calculate the water energy of the hydropower station, unit selection and actual operation, and integrates the influencing factors of the water energy of the hydropower station (number of units started, turbine characteristics, generator characteristics, head loss, etc. ) establishes a mathematical model, inputs the hydrological process parameters of the power station, simulates the basic conditions of the power station during normal operation and generates electricity, and cooperates with historical records to calculate accurate power generation results through weighted algorithms to guide the selection of hydropower stations, which can effectively The combination of hydrometeorological conditions and unit power generation characteristics simulates real-time efficiency, improves accuracy, and avoids inconsistencies between unit characteristics and hydrological conditions.

Although the present invention has been described in detail with general descriptions and specific examples above, it is obvious to those skilled in the art that some modifications or improvements can be made on the basis of the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.

Claims (6)

1. An algorithm for dynamic simulation of water energy in hydropower stations, characterized by including the following steps: S1. Online collection of the natural working conditions parameters of the operation of the hydrogenerator unit: head loss H, unit reference flow rate Q, runner diameter D; S2. Analyze the comprehensive characteristic curve of the hydraulic turbine model runner using the neural network to collect data and form a database; S3. Collect the relationship curve data and output limit line data reflected in the comprehensive characteristic curve diagram of the hydraulic turbine model runner, and import the collected data into the database; S4. Based on the turbine speed n and combined with the collected parameters in step S1, establish an objective function of unit operating condition efficiency C with unit flow rate A and unit speed B as parameters. According to the objective function and safety and stability constraints It is required to perform optimization with the database established by the neural network to obtain the optimal solution E ₁ = (A, B); S5. Derive [Q _i (t), H _i (t)] at each time point in the long historical sequence of the unit, determine the diameter and rotation speed (D _j , n _j ) of the unit, and depict the comprehensive characteristic curve of the turbine model runner Figure, use neural network to collect data for analysis; S6. Combine the invariants during the operation of a group of units with the hydrological process of the power station and reservoir dispatch control by using the dynamic simulation method. The dynamic simulation method is to put the invariants during the operation into a specific sequence of hydrological processes. Try to select a "satisfactory solution" from the comprehensive analysis of annual power generation, unit stability, cavitation performance, unit cost, and auxiliary facility cost; use neural networks to collect data sets for analysis, and analyze the water flow and flow rate of the hydropower station over the years. The upstream and downstream water level hydrological processes are input into the parameters of the turbine model, and the maximum power generation is used as the objective function to dynamically simulate the flow of each unit to make full use of water energy, and reselect (D _j ,n _j ) _m , the multi-year average power generation E ₂ =(A ₂ ,B ₂ ); S7. By optimizing the operating efficiency C of the unit, the requirements for safety and stability constraints and the established neural network database, the optimal solution E ₁ = (A, B) is obtained and the maximum power generation is used as the objective function. The multi-year average power generation E ₂ =(A ₂ ,B ₂ ) obtained by analyzing the neural network is finally used to solve E ₁ and E ₂ using a weighted algorithm to obtain the optimal solution E ^* . 2. An algorithm for dynamic simulation of water energy in hydropower stations according to claim 1, characterized in that, in step S4, the calculation formula of the unit flow rate A is: The calculation formula of the unit speed B is: 3. An algorithm for dynamic simulation of water energy in hydropower stations according to claim 2, characterized in that, in step S4, the water energy calculation simulation of hydropower stations adopts the following model: In the water energy calculation simulation model of hydropower stations, Among them: H _1(t) is the water head corresponding to the upstream at time t, H _2(t) is the water head corresponding to the downstream at time t, ΔH is the head difference between the upstream and downstream, V ₀ is the initial water volume of the hydropower station, V(t) is the volume of water corresponding to the hydropower station at time t, β is the blade opening, β _max is the maximum blade opening, eta is the turbine efficiency, η _i is the generator efficiency, ∑ _t Q ⁱ is the diversion flow of all turbines, Q _max is the maximum overflow of the turbine, α is the guide vane opening, α _max is the maximum guide vane opening; D is the turbine efficiency set. 4. An algorithm for dynamic simulation of water energy in hydropower stations according to claim 1, characterized in that, in step S5, the output calculation model in the hydropower station is as follows: Among them: P(t) is the maximum output of the turbine at time t, D _i is the diameter of the i-th turbine, H is the water head, α is the guide vane opening, β is the blade opening, Q _i is the unit of the i-th turbine Flow rate, n _i is the unit speed of the i-th hydraulic turbine, and eta is the generator efficiency. 5. An algorithm for dynamic simulation of water energy in hydropower stations according to claim 1, characterized in that, in step S6, the calculation model of power generation is: 6. An algorithm for dynamic simulation of water energy in hydropower stations according to claim 1, characterized in that, in step S7, the calculation formula of the optimal solution E ^* is as follows: Among them, E ^* =[(kA ₁ +rA ₂ ), (kB ₁ +rB ₂ )].