Method for selecting water diversion system model during frequency response of pumped storage unit
By analyzing the influence of key parameters θ and R1 on the water strike model, we simulate the frequency response characteristics of different water strike models, and solve the problem of insufficient selection of water diversion system models in the existing technology, providing a more comprehensive model selection method, and improving the Accuracy and computational efficiency of frequency response analysis.
Patent Information
- Application Number
- CN202411119460.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-26
AI Technical Summary
In the existing research on frequency response of hydropower stations, the water diversion system model is insufficient. The main problem is that it only studies a single key parameter, ignores the approximate elastic water strike model, and only compares the differences between rigid and elastic water strike models, and fails to Take into account the differences in the results of multiple frequency responses under different parameters.
By analyzing the effects of key parameters θ=Tr/Tw and R1=Tw/Ta on the model accuracy of the water diversion system, we simulate the lowest frequency point, quasi-steady state frequency, the lowest frequency response time, and the initial frequency drop rate of different water hit models. and compare the mean square error of the frequency response curve, determine the accuracy changes of different water hit models under different parameters, and select appropriate models to balance the calculation accuracy and complexity.
It provides a more comprehensive method for selecting water diversion system models, with more comprehensive research objects and parameters, and more accurate comparison results. It is suitable for frequency response analysis of different pumped storage units, improving the accuracy of model selection and calculation efficiency.
Smart Images

Figure CN119026509A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of pumped storage frequency response modeling, and relates to a method for selecting a water diversion system model during the frequency response of a pumped storage unit. Background Art
[0002] In the water diversion system module, water hammer effect is divided into two structures: elastic water hammer and rigid water hammer. Elastic water hammer takes into account complex fluid mechanics formulas and is closer to reality. Rigid water hammer ignores some losses in the pipeline and is widely used due to its simple calculation.
[0003] In general studies on frequency response of hydropower stations, the following three models are usually used for water diversion pipelines.
[0004] (1) Elastic water hammer model
[0005] Water hammer transfer function in the water diversion pipeline:
[0006] G h (S)=-2h w th(0.5T r S)(1)
[0007] Where: T r =2L / a is the water hammer phase, a is the water hammer wave velocity, L is the length of the water diversion pipeline; h w =T w / T r , T w is the water flow inertia time constant, h w Represents the characteristic parameters of the pressurized pipe section.
[0008] The hyperbolic tangent function is not conducive to analytical calculation, so the Taylor series is used to expand it into a polynomial:
[0009]
[0010] Let i = 1, and the elastic water hammer model is as shown in formula (3):
[0011]
[0012] (2) Rigid water hammer model
[0013] Let i = 0 in equation (2), and the rigid water hammer model is shown in equation (4):
[0014] G h (S)=-T w S(4)
[0015] (3) Approximate elastic water hammer model
[0016] In some studies, the elastic water hammer model is simplified and the cubic term in equation (3) is ignored, and the approximate elastic water hammer model can be obtained as follows:
[0017]
[0018] Among these three types of water hammer models, the elastic water hammer model has the highest calculation accuracy but is the most complex to calculate, the rigid water hammer model has the lowest calculation accuracy but is the simplest to calculate, and the calculation accuracy and complexity of the approximate elastic water hammer model are between the two.
[0019] To study the working characteristics of pumped storage power stations when participating in frequency regulation, the selection of water diversion system models is crucial. Therefore, it is necessary to analyze the physical process of pumped storage units participating in frequency response, analyze the influence of factors such as head fluctuation, pipeline rigidity, and water hammer effect on the water diversion system module, balance the model's requirements for calculation accuracy and speed, and select a reasonable pumped storage unit water diversion system model.
[0020] Regarding the influence of different water hammer models on the dynamic characteristics of unit frequency response, existing studies mainly compare the changes in unit frequency response characteristics when different water hammer models are used by changing key parameters. However, most of these studies have the following problems: (1) only the influence of a single key parameter is studied; (2) only the influence of different models on the stability domain and dynamic quality of frequency response is analyzed; (3) only the difference between rigid water hammer model and elastic water hammer model is compared, and the approximate elastic water hammer model is ignored.
[0021] In summary, it is necessary to consider the differences in the frequency response results of various water hammer models under different parameter conditions and propose a more reasonable and comprehensive method for selecting a water diversion system model. Summary of the invention
[0022] The present invention mainly improves the shortcomings of the above-mentioned water diversion system selection method and proposes a more comprehensive water diversion system model selection method.
[0023] In order to achieve the above object, the present invention adopts the following technical solution:
[0024] A method for selecting a water diversion system model when a pumped storage unit responds to frequency. The selection method analyzes two key parameters θ=T r / T w With R1 = T w / T a The influence on the accuracy of the water diversion system model is: θ represents the ratio of the water hammer phase length to the water flow inertia time constant, R1 represents the inertia ratio of the turbine generator set, T r Indicates that water hammer is progressive, T w represents the water flow inertia time constant, T aRepresents the inertia time constant of the unit. In the study of frequency response, the quasi-steady state and lowest point of frequency and the rate of decrease of accident frequency are important research parameters; at the same time, to study the degree of fit between the model and the actual situation, it is necessary to continuously sample the frequency response curve and calculate the difference between the model being studied and the actual situation. Therefore, the present invention studies the differences among the lowest point of frequency, quasi-steady-state frequency, time to reach the lowest point of frequency response, and initial frequency decrease rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model through simulation, and compares the mean square error of the frequency response curve to analyze the changes in the accuracy of different water hammer models under different parameters. The specific implementation process is as follows: Figure 1 As shown. Specifically including the following steps:
[0025] Step S1: Analyze the differences between the transfer functions of different water hammer models and obtain key influencing parameters; specifically:
[0026] According to the transfer functions of the three types of water hammer models, namely, equations (3), (4) and (5), the key influencing parameter is the water flow inertia time constant T w T r Therefore, it is necessary to set different T w With T r The size and proportion of T can be used to simulate the actual conditions of different pipe lengths and pipe wall elasticity, so as to study the differences between different water hammer models. w With T r , it is difficult to select a suitable numerical range to generate a frequency response curve suitable for research, and the proportional relationship requires additional calculation to obtain. At the same time, considering the inertia time constant T of the pumped storage unit a It is usually a fixed value that is difficult to change. The present invention sets the key parameter θ = T r / T w With R1 = T w / T a , by changing θ to change T w With T r The proportional relationship between θ and R1 is changed to change T w With T r The numerical size of .
[0027] Step S2: Set appropriate parameters to express the ratio and size relationship between key influencing parameters for simulation; specifically: Step S2-1: Change key parameters θ and R1 to simulate the water diversion system under different pipe lengths, pipe wall elasticity, etc. Figure 2The simulation model shown obtains the differences between the lowest frequency point, quasi-steady-state frequency, time to reach the lowest frequency response point, and initial frequency drop rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model, and compares the mean square error of the frequency response curve to analyze the changes in the accuracy of different water hammer models under different parameters. Considering that the elastic water hammer model has the highest accuracy among the three commonly used water hammer models, the present invention uses the frequency response results of the elastic water hammer model to represent the actual situation. If the rigid water hammer model and the approximate elastic water hammer model are too different from the elastic water hammer model under a certain parameter, the elastic water hammer model should be selected to ensure the accuracy of the model; otherwise, the rigid or approximate elastic model can be selected to reduce the calculation complexity.
[0028] Step S2-2: according to the simulation results, obtaining the error variation trend of the water hammer model under different parameters;
[0029] Step S3: Determine the applicable scope of different water hammer models according to the parameter range of common pumped storage units.
[0030] The beneficial effects of the present invention are:
[0031] The present invention studies the differences among the lowest frequency point, quasi-steady-state frequency, time to reach the lowest frequency response point, and initial frequency drop rate under different parameters of different water hammer models, and compares the mean square error of the frequency response curves. Compared with previous studies, the research objects, parameters, and final comparison results are more comprehensive, which provides a new idea for the selection of water diversion system models. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is the overall structure designed by the present invention;
[0033] Figure 2 It is a Matlab / Simulink simulation model;
[0034] Figure 3 is the lowest frequency point of different water hammer models (Hz); Figure 3 (a) The lowest frequency point of different water hammer models under different R1 conditions when θ = 0.2; Figure 3 (b) The lowest frequency points of different water hammer models under different R1 conditions when θ = 3; Figure 3 (c) is the lowest frequency point of different water hammer models under different θ conditions with R1 = 0.1; Figure 3 (d) is the lowest frequency point of different water hammer models under different θ conditions with R1 = 0.4;
[0035] Figure 4 is the quasi-steady-state frequency of different water hammer models (Hz); Figure 4 (a) Quasi-steady-state frequencies of different water hammer models under different R1 conditions when θ = 0.2; Figure 4(b) Quasi-steady-state frequencies of different water hammer models under different R1 conditions when θ = 3; Figure 4 (c) Quasi-steady-state frequencies of different water hammer models under different θ conditions with R1 = 0.1; Figure 4 (d) Quasi-steady-state frequencies of different water hammer models under different θ conditions with R1 = 0.4;
[0036] Figure 5 is the mean square error of the frequency response curve of the rigid, approximately elastic water hammer model; Figure 5 (a) The mean square error of the frequency response curves of different water hammer models under different R1 conditions when θ = 0.2; Figure 5 (b) is the mean square error of the frequency response curves of different water hammer models under different R1 conditions when θ=3; Figure 5 (c) is the mean square error of the frequency response curves of different water hammer models under different θ conditions with R1 = 0.1; Figure 5 (d) is the mean square error of the frequency response curves of different water hammer models under different θ conditions with R1 = 0.4;
[0037] Figure 6 is the time (s) for different water hammer models to reach the lowest frequency point; Figure 6 (a) The time for different water hammer models to reach the lowest frequency point under different R1 conditions when θ = 0.2; Figure 6 (b) The time for different water hammer models to reach the lowest frequency point when θ = 3 and R1 is different; Figure 6 (c) The time for different water hammer models to reach the lowest frequency point under different θ conditions with R1 = 0.1; Figure 6 (d) The time for different water hammer models to reach the lowest frequency point under different θ conditions with R1 = 0.4;
[0038] Figure 7 is the initial frequency drop rate of different water hammer models (Hz / s); Figure 7 (a) The initial frequency drop rate of different water hammer models under different R1 conditions when θ = 0.2; Figure 7 (b) The initial frequency drop rate of different water hammer models under different R1 conditions when θ = 3; Figure 7 (c) The initial frequency drop rate of different water hammer models under different θ conditions with R1 = 0.1; Figure 7 (d) is the initial frequency drop rate of different water hammer models under different θ conditions with R1=0.4. Specific implementation methods
[0040] As a clean energy source, pumped storage units have the advantages of high energy conversion rate, rapid startup, flexible and reliable operation, etc., and are important frequency response resources. When studying the dynamic frequency response model of pumped storage units, the operating point often deviates from the rated operating conditions, and the operating parameters change accordingly, resulting in the static model transfer function of the water diversion system no longer being applicable. Therefore, it is necessary to analyze the physical process of the pumped storage unit participating in the frequency response, analyze the influence of factors such as head fluctuations, pipeline rigidity, and water hammer effect on the water diversion system module, balance the model's requirements for calculation accuracy and calculation speed, and select a reasonable pumped storage unit water diversion system model. Therefore, the present invention mainly improves the shortcomings of the above-mentioned water diversion system selection method, and proposes a more comprehensive water diversion system model selection method.
[0041] In order to achieve the above object, the present invention adopts the following technical solution:
[0042] A method for selecting a water diversion system model when a pumped storage unit responds to frequency. The selection method analyzes two key parameters θ=T r / T w With R1 = T w / T a The influence on the accuracy of the water diversion system model is: θ represents the ratio of the water hammer phase length to the water flow inertia time constant, R1 represents the inertia ratio of the turbine generator set, T r Indicates that water hammer is progressive, T w represents the water flow inertia time constant, T a Represents the inertia time constant of the unit. In the study of frequency response, the quasi-steady state and lowest point of frequency and the rate of decrease of accident frequency are important research parameters; at the same time, to study the consistency between the model and the actual situation, it is necessary to continuously sample the frequency response curve and calculate the difference between the model being studied and the actual situation. Therefore, the present invention studies the differences among the lowest point of frequency, quasi-steady-state frequency, time to reach the lowest point of frequency response, and initial frequency decrease rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model, and compares the mean square error of the frequency response curve, and analyzes the changes in the accuracy of different water hammer models under different parameters. The specific implementation process is as follows: Figure 1 As shown. Specifically including the following steps:
[0043] Step S1: Analyze the differences between the transfer functions of different water hammer models and obtain key influencing parameters; specifically:
[0044] According to the transfer functions of the three types of water hammer models, namely, equations (3), (4) and (5), the key influencing parameter is the water flow inertia time constant T w T r Therefore, it is necessary to set different T w With T rThe size and proportion of T can be studied to study the differences of different water hammer models. w With T r , it is difficult to select a suitable numerical range to generate a frequency response curve suitable for research, and the proportional relationship requires additional calculation to obtain. At the same time, considering the inertia time constant T of the pumped storage unit a It is usually a fixed value that is difficult to change, so the present invention selects T a Based on, set the key parameter θ = T r / T w With R1 = T w / T a , by changing θ to change T w With T r The proportional relationship between θ and R1 is changed to change T w With T r The numerical size of .
[0045] Step S2: Set appropriate parameters to express the ratio and size relationship between key influencing parameters for simulation; specifically: Step S2-1: Change the key parameters θ and R1 to simulate the water diversion system under different pipe lengths, pipe wall elasticity, etc. The differences between the lowest frequency point, quasi-steady-state frequency, time to reach the lowest frequency response point, and initial frequency drop rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model are obtained through simulation, and the mean square error of the frequency response curve is compared to analyze the changes in the accuracy of different water hammer models under different parameters. Considering that the elastic water hammer model has the highest accuracy among the three commonly used water hammer models, the present invention uses the frequency response results of the elastic water hammer model to represent the actual situation. If the rigid water hammer model and the approximate elastic water hammer model are too different from the elastic water machine model under a certain parameter, the elastic water hammer model should be selected to ensure the accuracy of the model; otherwise, a rigid or approximate elastic model can be selected to reduce the calculation complexity.
[0046] Step S2-2: Analyze the simulation results to obtain the error variation trend of the water hammer model under different parameters;
[0047] Compare Figures 3 to 7It can be obtained that: when R1=0.1 or θ=0.2, the initial frequency drop rate, the lowest frequency point and arrival time, and the quasi-steady-state frequency of the rigid water hammer model and the approximate elastic water hammer model are similar to those of the elastic water hammer model, and the mean square error of the frequency response curve is small; with the increase of θ and R1, the difference between the frequency response curves of the rigid water hammer model and the approximate elastic water hammer model compared with the elastic water hammer model shows an overall upward trend; when R1=0.4 and θ>1, compared with the case of R1=0.1 and θ=0.2, the difference between the lowest frequency response point of the rigid, approximate elastic water hammer model and the elastic water hammer model and the mean square error of the frequency response curve increase several times or even dozens of times. Similarly, when R1>0.2 and θ=3, compared with the case of R1=0.1 and θ=0.2, the difference between the lowest frequency response point of the rigid, approximate elastic water hammer model and the elastic water hammer model and the mean square error of the frequency response curve also increase several times or even dozens of times. It can be seen that when θ and R1 are small, the calculation accuracy of the three types of water hammer models is slightly different; as θ and R1 increase, the difference between the rigid water hammer model and the approximate elastic water hammer model compared with the elastic water hammer model shows an overall upward trend.
[0048] Step S3: Analyze the parameter range of common pumped storage units and determine the applicable range of different water hammer models.
[0049] At present, the water flow inertia time constant T of my country's pumped storage units is w The value is mostly between 1.5 and 2.5s. For some pumped storage power stations with higher speed regulation performance requirements, T w Less than 2s, and R1 less than 0.2; if the pumped storage power station accounts for a small proportion of the power grid, there is no need to adjust T w Considering the water hammer phase length T r It is positively correlated with the water head height. Therefore, for small pumped storage units, the length of their water diversion pipelines is often similar to that of ordinary hydropower units, so the range of θ is similar to that of ordinary hydropower units; while the water diversion pipelines of medium and large pumped storage units are often much longer than those of ordinary hydropower units, so the θ of medium and large units is usually much larger than that of ordinary hydropower units. In general, for pumped storage units with θ<1 and R1<0.2, a rigid water hammer model can be used to reduce the calculation complexity; for pumped storage units with θ>1 and R1>0.2, an elastic water hammer model should be used to improve the calculation accuracy; in other cases, an approximate elastic water hammer model can be used.
[0050] The above-described embodiments merely express the implementation methods of the present invention, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.
Claims
1. A method for selecting a water diversion system model during frequency response of a pumped storage unit, characterized in that: The selection method studies the differences among the lowest frequency point, quasi-steady-state frequency, time to reach the lowest frequency response point, and initial frequency drop rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model through simulation, and compares the mean square error of the frequency response curve, analyzes the change of the accuracy of different water hammer models under different parameters, and finally selects two key parameters θ=T r / T w With R1 = T w / T a The influence on the accuracy of the water diversion system model is: θ represents the ratio of the water hammer phase length to the water flow inertia time constant, R1 represents the inertia ratio of the turbine generator set, T r Indicates that water hammer is progressive, T w represents the water flow inertia time constant, T a Indicates the unit inertia time constant.
2. The method for selecting a water diversion system model when a pumped storage unit frequency responds according to claim 1 is characterized in that: The specific steps include: Step S1: Analyze the differences between the transfer functions of different water hammer models and obtain key influencing parameters; specifically: According to the transfer functions of the elastic water hammer model, the rigid water hammer model, and the approximate elastic water hammer model, the key influencing parameter is the water flow inertia time constant T. w T r ; Therefore, different T w With T r The size and proportion relationship of the 3D image can simulate the actual situation; Set the key parameter θ = T r / T w With R1 = T w / T a , by changing θ to change T w With T r The proportional relationship between θ and R1 is changed to change T w With T r The numerical value of Step S2: Setting appropriate parameters to express the ratio and size relationship between key influencing parameters for simulation; specifically: Step S2-1: Change the key parameters θ and R1 to simulate the water diversion system under different pipe lengths and pipe wall elasticity conditions; obtain the differences among the lowest frequency point, quasi-steady-state frequency, time to reach the lowest frequency response point, and initial frequency drop rate of the rigid water hammer model, the approximate elastic water hammer model, and the elastic water hammer model; compare the mean square error of the frequency response curves, and analyze the changes in the accuracy of different water hammer models under different parameters; use the frequency response results of the elastic water hammer model to represent the actual situation. If the rigid water hammer model and the approximate elastic water hammer model are too different from the elastic water hammer model under a certain parameter, the elastic water hammer model is selected; otherwise, the rigid or approximate elastic model can be selected; Step S2-2: according to the simulation results, obtaining the error variation trend of the water hammer model under different parameters; Step S3: Determine the applicable scope of different water hammer models according to the parameter range of common pumped storage units.
Citation Information
Cited By
Head wave position prediction method for water hammer effect of fragment impact liquid-filled container
CN120180108A
