Full-condition power angle prediction method and system for steam turbine generator and computer storage medium

Abstract

The application relates to a full-working-condition power angle prediction method and system of a steam turbine generator and a computer storage medium, the power angle prediction method comprising the following steps: obtaining historical working condition data of the generator, wherein the historical working condition data comprises multiple groups of detection data under multiple active power levels; determining optimal synchronous reactance corresponding to each active power level according to the historical working condition data; obtaining an active power level of a to-be-predicted working condition, and calculating synchronous reactance corresponding to the to-be-predicted working condition according to the optimal synchronous reactance corresponding to each active power level; obtaining active power, reactive power and machine terminal voltage of the generator under the to-be-predicted working condition, and calculating a power angle corresponding to the to-be-predicted working condition according to the active power, the reactive power, the machine terminal voltage and the synchronous reactance. Through the technical scheme, the power angle has high accuracy, and the safety and stability of unit operation are ensured.

Description

Technical Field

[0001] This invention relates to the field of power system stability, and in particular to a method, system, and computer storage medium for predicting the power angle of a steam turbine generator under all operating conditions. Background Technology

[0002] To address the imbalance between my country's energy endowment and economic development, and the mismatch between supply and demand, the voltage levels of power grids in various regions have gradually increased and become interconnected. With the continuous development of modern large-scale power grids and the large-scale use of cable lines, the equivalent capacitance to ground in the lines has increased. This capacitance rise effect has exacerbated the problems of reactive power excess and voltage rise at the line terminals, particularly during off-peak periods such as holidays and midnight. In response to the reality of reactive power excess, the power grid has placed higher demands on the leading-phase (reactive power absorption) voltage regulation capabilities of generator units.

[0003] Deeply advanced phase operation directly leads to an increase in the generator's power angle, resulting in reduced system stability. Exceeding the stability limit will cause the generator to lose synchronization, causing power fluctuations that trigger system oscillations and subsequently a chain reaction affecting other units. Due to the influence of nonlinear factors such as ferromagnetic material saturation and magnetic field distortion, existing methods cannot accurately calculate the power angle, leading to inaccurate assessment of the unit's safety margin by the control center and posing safety risks. Therefore, there is an urgent need to propose a power angle estimation method that can accurately predict the power angle before operation under various conditions, satisfying all operating conditions of the steam turbine generator. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, system and computer storage medium for predicting the power angle of a steam turbine generator under all operating conditions, in view of the above-mentioned defects in the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: constructing a method for predicting the power angle of a steam turbine generator under all operating conditions, including:

[0006] Step S10: Obtain historical operating condition data of the generator, wherein the historical operating condition data includes multiple sets of detection data under multiple active power levels, and each set of detection data includes the active power, reactive power, terminal voltage and power angle of the generator detected under the corresponding active power level.

[0007] Step S20: Based on the historical operating data, determine the optimal synchronous reactance corresponding to each active power level;

[0008] Step S30: Obtain the active power level of the operating condition to be predicted, and calculate the synchronous reactance corresponding to the operating condition to be predicted based on the optimal synchronous reactance corresponding to each active power level.

[0009] Step S40: Obtain the active power, reactive power, and terminal voltage of the generator under the predicted operating condition, and calculate the power angle corresponding to the predicted operating condition based on the active power, reactive power, terminal voltage, and synchronous reactance.

[0010] Preferably, step S20 includes:

[0011] Based on the historical operating data, the optimal synchronous reactance corresponding to each active power level is determined using the particle swarm optimization algorithm.

[0012] Preferably, step S30, which involves calculating the synchronous reactance corresponding to the predicted operating condition based on the optimal synchronous reactance corresponding to each active power level, includes:

[0013] Based on the optimal synchronous reactance corresponding to each active power level, the synchronous reactance corresponding to the predicted operating condition is calculated using interpolation.

[0014] Preferably, between step S30 and step S40, the method further includes:

[0015] Step S51: Obtain the unsaturated reactance from the generator's nameplate information;

[0016] Step S52: Determine whether the synchronous reactance calculated in step S30 deviates from the preset percentage of the unsaturated reactance. If yes, proceed to step S40; otherwise, proceed to step S53.

[0017] Step S53: Use the unsaturated reactance as the synchronous reactance corresponding to the operating condition to be predicted, and then execute step S40.

[0018] Preferably, in step S40, the power angle corresponding to the predicted operating condition is calculated according to the following formula:

[0019]

[0020] Where δ is the power angle corresponding to the operating condition to be predicted, P G Let Q be the active power under the operating condition to be predicted. G U represents the reactive power under the predicted operating condition. G X represents the terminal voltage under the predicted operating condition. d The synchronous reactance is the operating condition to be predicted.

[0021] Preferably, step S20 includes:

[0022] Step S21: Construct the fitness function of the synchronous reactance during the identification stage;

[0023] Step S22: Initialize the position and velocity of the particles in the particle swarm, wherein the position of the particle represents the value of the synchronization reactance;

[0024] Step S23: Calculate the fitness value corresponding to the current position of the particle according to the fitness function;

[0025] Step S24: Calculate the individual optimal position currently searched by the particle and the population optimal position currently searched by the particle swarm, and update the current position and current velocity of the particle based on the individual optimal position and the population optimal position.

[0026] Step S25: Determine whether the termination condition is met. If yes, proceed to step S26; otherwise, proceed to step S23.

[0027] Step S26: The current position of the particle is taken as the optimal synchronous reactance corresponding to the corresponding active power level.

[0028] Preferably, the step of determining whether the termination condition is currently met includes:

[0029] Determine if the maximum number of iterations has been reached; or,

[0030] Determine if the current fitness function meets the accuracy condition.

[0031] Preferably, in step S30, the synchronous reactance corresponding to the operating condition to be predicted is calculated according to the following formula:

[0032]

[0033] in, P is the synchronous reactance corresponding to the operating condition to be predicted. p P represents the active power level corresponding to the operating condition to be predicted. k P k+1 These are the two active power levels adjacent to the active power level corresponding to the operating condition to be predicted. It represents the optimal synchronous reactance corresponding to the two adjacent active power levels.

[0034] The present invention also constructs a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for predicting the power angle of a steam turbine generator under all operating conditions.

[0035] The present invention also constructs a steam turbine generator full-condition power angle prediction system, including a processor and a memory storing a computer program. When the processor executes the computer program, it implements the steps of the steam turbine generator full-condition power angle prediction method described above.

[0036] In the technical solution provided by this invention, the mechanism-data hybrid driving method can effectively overcome the influence of nonlinear factors on the accuracy of power angle calculation. Therefore, the power angle accuracy is high, which provides a guarantee for the safe and stable operation of the unit. Moreover, the power angle calculation results of the turbine generator under all operating conditions can be obtained by repeatedly executing the above scheme. Attached Figure Description

[0037] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0038] Figure 1A This is a graph showing the measured data of the synchronous reactance and active power of an existing generator.

[0039] Figure 1B This is a graph showing the measured data of synchronous reactance and active power of another existing generator;

[0040] Figure 2 This is a flowchart of Embodiment 1 of the steam turbine generator full-condition power angle prediction method of the present invention;

[0041] Figure 3 yes Figure 2 The flowchart of step S20 in Example 1. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Firstly, it should be noted that, based on the fundamental theory and phasor diagram of salient-pole generators, the calculation of the generator's power angle can be transformed into the calculation of synchronous reactance. However, during generator operation, the synchronous reactance is difficult to accurately obtain due to the influence of nonlinear factors, leading to significant errors in existing calculation methods. Based on actual field data, such as... Figure 1A , Figure 1B As shown, under the same power level (especially active power level), the synchronous reactance of the generator does not vary significantly and exhibits a linear relationship with the power level. Therefore, based on this premise, a mechanism-plus-data hybrid modeling approach can be adopted to construct a mathematical model for estimating the synchronous reactance and power angle of the generator. This approach combines electrical mechanism analysis with historical data mining to improve the accuracy of the calculations.

[0044] Figure 2 This is a flowchart of an embodiment of the turbine generator full-condition power angle prediction method of the present invention. The turbine generator full-condition power angle prediction method of this embodiment includes the following steps:

[0045] Step S10: Obtain historical operating condition data of the generator, wherein the historical operating condition data includes multiple sets of detection data under multiple active power levels, and each set of detection data includes the active power, reactive power, terminal voltage and power angle of the generator detected under the corresponding active power level.

[0046] In this step, the distributed control system (DCS) of the power plant records multiple sets of active power, reactive power, generator terminal voltage, and power angle data. Therefore, historical operating condition data can be obtained from the DCS, and this historical operating condition data, for example, includes n sets of detection data under m active power platforms to form a data matrix.

[0047] Step S20: Based on the historical operating data, determine the optimal synchronous reactance corresponding to each active power level;

[0048] In this step, the particle swarm optimization algorithm is preferably used to determine the optimal synchronous reactance corresponding to each active power level.

[0049] Step S30: Obtain the active power level of the operating condition to be predicted, and calculate the synchronous reactance corresponding to the operating condition to be predicted based on the optimal synchronous reactance corresponding to each active power level.

[0050] In this step, based on the optimal synchronous reactance corresponding to multiple determined active power levels, a mechanistic model is used to calculate the different reactances for the operating condition to be predicted. Furthermore, interpolation is preferably used to calculate the synchronous reactance corresponding to the operating condition to be predicted.

[0051] Step S40: Obtain the active power, reactive power, and terminal voltage of the generator under the predicted operating condition, and calculate the power angle corresponding to the predicted operating condition based on the active power, reactive power, terminal voltage, and synchronous reactance.

[0052] This embodiment's technical solution acquires multiple sets of active power, reactive power, terminal voltage, and power angle data for known active power levels using historical generator operating data, thereby determining the optimal synchronous reactance corresponding to each active power level. Then, combining the determined optimal synchronous reactance for each active power level, the synchronous reactance corresponding to the active power level under the predicted operating condition is determined. Finally, based on the determined synchronous reactance and the active power, reactive power, and terminal voltage under the predicted operating condition, the power angle corresponding to the predicted operating condition can be calculated. This power angle calculation method employs a mechanism-data hybrid driving approach, effectively overcoming the influence of nonlinear factors on the accuracy of power angle calculation. Therefore, the power angle accuracy is high, providing a guarantee for the safe and stable operation of the unit. Furthermore, by repeatedly executing the above scheme, the power angle calculation results for the turbine generator under all operating conditions can be obtained.

[0053] Furthermore, based on the fundamental theory of salient-pole synchronous generators and their phasor diagrams, we can obtain:

[0054]

[0055] Where δ is the power angle of the generator at the corresponding active power level, P G Q represents the active power of the generator at the corresponding active power level. G U represents the reactive power at the corresponding active power level. G X represents the terminal voltage at the corresponding active power level. d For the corresponding active power level, I is the synchronous reactance. G This refers to the terminal current at the corresponding active power level. This refers to the power factor at the corresponding active power level.

[0056] Considering Multiply the numerator and denominator on the right side of Equation 1 by U. G After a simple derivation, the formula for calculating the generator's power angle is shown in Formula 2:

[0057]

[0058] Therefore, in a specific embodiment, in step S40, the power angle corresponding to the predicted operating condition can be calculated according to the above formula 2. It should be understood that when using formula 2 for calculation, P G Q G U G X d These are the active power, reactive power, terminal voltage, and synchronous reactance under the operating conditions to be predicted.

[0059] Furthermore, in an optional embodiment, when using the particle swarm optimization algorithm to determine the optimal synchronization reactance corresponding to each active power level, combined with Figure 3 Step S20 specifically includes:

[0060] Step S21: Construct the fitness function of the synchronous reactance during the identification stage;

[0061] Step S22: Initialize the position and velocity of the particles in the particle swarm, wherein the position of the particle represents the value of the synchronization reactance;

[0062] Step S23: Calculate the fitness value corresponding to the current position of the particle according to the fitness function;

[0063] Step S24: Calculate the individual optimal position currently searched by the particle and the population optimal position currently searched by the particle swarm, and update the current position and current velocity of the particle based on the individual optimal position and the population optimal position.

[0064] Step S25: Determine whether the termination condition is met. If yes, proceed to step S26; otherwise, proceed to step S23.

[0065] In this step, the termination condition can be determined by the following criteria: whether the maximum number of iterations has been reached; or, whether the current fitness function meets the accuracy condition.

[0066] Step S26: The current position of the particle is taken as the optimal synchronous reactance corresponding to the corresponding active power level.

[0067] In one specific embodiment, the fitness function of the constructed synchronous reactance during the identification phase is shown in the following formula:

[0068]

[0069]

[0070] In the formula, Let J be the synchronous reactance to be corrected under the active power level of the j-th group; The power angle measurement value in the i-th group of detection data under the j-th group of active power level; P is the calculated power angle value corresponding to the i-th group of detection data under the j-th group of active power level; i j , These represent the active power measurement value, reactive power measurement value, and terminal voltage measurement value in the i-th group of detection data under the j-th group of active power level.

[0071] The mathematical model for identifying synchronous reactance parameters, constructed with minimizing the fitness function as the optimization objective, is shown below:

[0072]

[0073] Where, in the formula, The unsaturated reactance is from the generator's nameplate information, and the synchronous reactance obtained in the optimization model is set to meet a deviation error of ±50%.

[0074] When using the particle swarm optimization algorithm in swarm optimization, the specific solution process is as follows: For a spatial problem with a feasible region of d-dimensional real numbers, since the particle swarm needs to continuously update its position and velocity information, the particle's position X and velocity V are set in the feasible region, corresponding to the feasible solution of the actual problem and the amount that the feasible solution needs to be adjusted during the iteration process, respectively. During the iteration process, the update formulas for the particle's position and velocity are as follows:

[0075]

[0076]

[0077] In the formula, x is the position of the corresponding particle, representing the synchronization reactance to be corrected; v is the velocity of the corresponding particle; i is the particle number, i = 1, 2, 3, ..., l; l is the population size; k is the number of iterations; p b This represents the particle's current optimal position (i.e., its individual optimal position); g b ω represents the current optimal position found by the particle swarm (population optimal position); ω represents the velocity inertia weight; c1 and c2 are learning factors, reflecting the weights of self-summary and population learning on the search process, respectively; r1 and r2 are random numbers in the interval [0,1].

[0078] Formula 7 consists of three parts: As an inertial component, each flying particle has a tendency to maintain its original state of motion; As an individual cognitive component, the current position is adjusted using the individual optimal position of the current particle, reflecting the influence of individual experience in the algorithm model; As a component of social cognition, it reflects the impact of overall information.

[0079] The quality of a particle is measured using the fitness function defined in Equation 3. After each position update, the corresponding fitness function value is calculated and compared with the previous individual best value and the population best position, thus updating p. b and g b The formula is as follows:

[0080]

[0081]

[0082] When the fitness function meets the accuracy condition or the algorithm reaches the maximum number of iterations, the optimal solution to the problem can be obtained, that is, the optimal synchronous reactance under the active power level.

[0083] Furthermore, in an optional embodiment, step S20 can obtain the optimal synchronization reactance corresponding to each of the m groups of active power levels, and form an optimal synchronization reactance sequence. Once the active power level of the operating condition to be predicted is determined, the synchronous reactance corresponding to the operating condition to be predicted can be calculated using linear interpolation. Step S30 can specifically calculate the synchronous reactance corresponding to the operating condition to be predicted according to Formula 10:

[0084]

[0085] in, P is the synchronous reactance corresponding to the operating condition to be predicted. p P represents the active power level corresponding to the operating condition to be predicted. k P k+1 These are the two active power levels adjacent to the active power level corresponding to the operating condition to be predicted. These are the optimal synchronous reactances corresponding to the two adjacent active power levels, respectively.

[0086] Furthermore, in an optional embodiment, between step S30 and step S40, the following step is further included:

[0087] Step S51: Obtain the unsaturated reactance from the generator's nameplate information;

[0088] In this step, the generator's nameplate information may include the following parameters: rated apparent capacity, rated voltage, and unsaturated reactance. Therefore, the generator's unsaturated reactance can be obtained by reading the generator's nameplate information.

[0089] Step S52: Determine whether the synchronous reactance calculated in step S30 deviates from the preset percentage of the unsaturated reactance. If yes, proceed to step S40; otherwise, proceed to step S53.

[0090] Step S53: Use the unsaturated reactance as the synchronous reactance corresponding to the operating condition to be predicted, and then execute step S40.

[0091] In this embodiment, after calculating the synchronous reactance corresponding to the predicted operating condition in step S30, to prevent overfitting of the synchronous reactance result due to calculation errors in reactance identification, synchronous reactances within a preset percentage (e.g., 5%) of the unsaturated reactance can be set as unsaturated reactances. Only when the calculated synchronous reactance deviates from the preset percentage of the unsaturated reactance is the calculation result used as the synchronous reactance corresponding to the predicted operating condition. Specifically, Formula 10 can be optimized as follows:

[0092]

[0093] The following example uses a megawatt-class nuclear power unit to estimate the generator power angle according to the method described in the above embodiments, and compares it with experimental values ​​to verify the feasibility and accuracy of the power angle prediction method of the present invention:

[0094] First, the generator nameplate information is shown in Table 1, and the collected historical operating condition data is shown in Table 2.

[0095] Table 1

[0096]

[0097] Table 2

[0098]

[0099] According to Table 2, the historical operating condition data consists of four sets of detection data each for two active power levels: 1060MW and 860MW. We assume the operating condition to be predicted is a 1000MW active power level. First, using the historical operating condition data in Table 2 as input, mathematical models of the synchronous reactance under the 1060MW and 860MW active power levels are established respectively. Particle swarm optimization algorithm is used for parameter identification to obtain the synchronous reactance identification results for the above operating conditions. Then, based on the synchronous reactance identified under the 1060MW and 860MW active power levels, the synchronous reactance of the generator under the 1000MW active power platform is obtained through linear interpolation, thus obtaining the calculated power angle value. The calculated power angle value is compared with the actual measured power angle value, and the results are shown in Table 3.

[0100] Table 3

[0101]

[0102] As can be seen from Table 3, under different operating conditions, the power angle calculated by the power angle prediction method in the above embodiment is close to the actual measured value, with a maximum relative error of less than 0.7%.

[0103] The present invention also constructs a computer storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for predicting the power angle of a steam turbine generator under all operating conditions.

[0104] The present invention also constructs a steam turbine generator full-condition power angle prediction system, which includes a processor and a memory storing a computer program. When the processor executes the computer program, it implements the steps of the steam turbine generator full-condition power angle prediction method described above.

[0105] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any alterations, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for predicting the power angle of a steam turbine generator under all operating conditions, characterized in that, include: Step S10: Obtain historical operating condition data of the generator, wherein the historical operating condition data includes multiple sets of detection data under multiple active power levels, and each set of detection data includes the active power, reactive power, terminal voltage and power angle of the generator detected under the corresponding active power level. Step S20: Based on the historical operating data, determine the optimal synchronous reactance corresponding to each active power level; Step S30: Obtain the active power level of the operating condition to be predicted, and calculate the synchronous reactance corresponding to the operating condition to be predicted based on the optimal synchronous reactance corresponding to each active power level. Step S40: Obtain the active power, reactive power, and terminal voltage of the generator under the predicted operating condition, and calculate the power angle corresponding to the predicted operating condition based on the active power, reactive power, terminal voltage, and synchronous reactance. Step S30, which involves calculating the synchronous reactance corresponding to the predicted operating condition based on the optimal synchronous reactance for each active power level, includes the following steps: Based on the optimal synchronous reactance corresponding to each active power level, the synchronous reactance corresponding to the predicted operating condition is calculated using interpolation.

2. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 1, characterized in that, Step S20 includes: Based on the historical operating data, the optimal synchronous reactance corresponding to each active power level is determined using the particle swarm optimization algorithm.

3. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 1, characterized in that, Between step S30 and step S40, the following is also included: Step S51: Obtain the unsaturated reactance from the generator's nameplate information; Step S52: Determine whether the synchronous reactance calculated in step S30 deviates from the preset percentage of the unsaturated reactance. If yes, proceed to step S40; otherwise, proceed to step S53. Step S53: Use the unsaturated reactance as the synchronous reactance corresponding to the operating condition to be predicted, and then execute step S40.

4. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 1, characterized in that, In step S40, the power angle corresponding to the predicted operating condition is calculated according to the following formula: in, The power angle corresponding to the operating condition to be predicted. The active power under the operating condition to be predicted is... The reactive power under the predicted operating condition is... The terminal voltage under the predicted operating condition is... The synchronous reactance is the operating condition to be predicted.

5. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 2, characterized in that, Step S20 includes: Step S21: Construct the fitness function of the synchronous reactance during the identification stage; Step S22: Initialize the position and velocity of the particles in the particle swarm, wherein the position of the particle represents the value of the synchronization reactance; Step S23: Calculate the fitness value corresponding to the current position of the particle according to the fitness function; Step S24: Calculate the individual optimal position currently searched by the particle and the population optimal position currently searched by the particle swarm, and update the current position and current velocity of the particle based on the individual optimal position and the population optimal position. Step S25: Determine whether the termination condition is met. If yes, proceed to step S26; otherwise, proceed to step S23. Step S26: The current position of the particle is taken as the optimal synchronous reactance corresponding to the corresponding active power level.

6. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 5, characterized in that, The step of determining whether the termination condition is currently met includes: Determine if the maximum number of iterations has been reached; or, Determine if the current fitness function meets the accuracy condition.

7. The method for predicting the power angle of a steam turbine generator under all operating conditions according to claim 1, characterized in that, In step S30, the synchronous reactance corresponding to the operating condition to be predicted is calculated according to the following formula: in, The synchronous reactance corresponding to the operating condition to be predicted. The active power level corresponding to the operating condition to be predicted. , These are the two active power levels adjacent to the active power level corresponding to the operating condition to be predicted. , It represents the optimal synchronous reactance corresponding to the two adjacent active power levels.

8. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the full-condition power angle prediction method for steam turbine generators according to any one of claims 1-7.

9. A steam turbine generator full-condition power angle prediction system, comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the full-condition power angle prediction method for steam turbine generators according to any one of claims 1-7.

Images