Frequency response aggregation model of multi-gas turbine unit system based on weighted reduced order

Abstract

The present application belongs to the technical field of power grid frequency stability analysis, and particularly relates to a frequency response aggregation model of a multi-gas turbine unit system based on weighted reduction. The present application comprises: step 1, factor decomposition of a speed regulator part; step 2, model aggregation in the speed regulation system part of a generator unit; step 3, error derivation of the speed regulator aggregation link; and step 4, obtaining the frequency response aggregation model of the multi-gas turbine unit system based on weighted reduction based on the above three steps. The present application can realize a simplified model of a frequency dynamic process, and the built model can reduce the order of the frequency response model of the multi-gas turbine unit system through simplification and reduction in the thought of weighted average, realize fast analysis of the frequency response process of the system, and has certain advantages in solving the dynamic characteristics of the model.

Description

Technical Field

[0001] This invention belongs to the field of power grid frequency stability analysis technology, and in particular relates to a frequency response aggregation model for a multi-gas turbine system based on weighted order reduction. Background Technology

[0002] In the dynamic process analysis of power system frequency response, the maximum frequency difference is a key parameter for frequency monitoring and an important basis for frequency control decisions, making its solution method particularly important. It is mainly divided into time-series simulation method and analytical method.

[0003] The timing simulation method has high accuracy in solving system frequency calculations, but as the number of system units increases and the overall system scale expands, its order also increases, which will significantly increase the simulation time and cause a significant decrease in solution efficiency.

[0004] Analytical methods are fast at calculating system frequencies, but their accuracy is not as high as simulation methods. Existing analytical methods lack characterization of gas turbines, so it is necessary to rationally reduce the order of the model to improve the solution efficiency.

[0005] Currently, there are aggregate models based on multiple thermal power units, which can effectively improve the solution speed of the system. However, there is no relevant research on the derivation and construction of aggregate models for multiple gas turbine systems. Summary of the Invention

[0006] To address the shortcomings of the existing technologies, this invention provides a frequency response aggregation model for multi-gas turbine systems based on weighted order reduction. Its purpose is to achieve a simplified model of the frequency dynamic process.

[0007] The technical solution adopted by the present invention to achieve the above objectives is as follows:

[0008] The frequency response aggregation model of a multi-gas turbine system based on weighted order reduction includes the following steps:

[0009] Step 1. Factorize the speed controller section;

[0010] Step 2. Perform model aggregation on the speed control system of the generator set;

[0011] Step 3. Derivation of the error in the speed governor's convergence stage;

[0012] Step 4. Based on the above three steps, obtain the frequency response aggregation model of the multi-gas turbine system based on weighted order reduction.

[0013] Furthermore, the factorization of the governor section described in step 1 is based on the control principle of the gas turbine unit, and establishes a system frequency response model including the governor, valve positioner, fuel system and combustion chamber and generator section.

[0014] Define R G K is the system's droop coefficient. G c represents the power capacity ratio. g and b g X represents the coefficients of the transfer function of the valve positioner in the fuel system. g It is the lead time constant of the speed governor section, Y g It is the lag time constant of the speed governor section, T CR It is the combustion reaction delay time constant, T F It is the fuel system time constant, T CD It is the compression displacement time constant;

[0015] Based on the concept of unit aggregation model, the parameters of the frequency response model of multi-gas turbine unit system are aggregated to form a frequency response aggregation model of multi-gas turbine unit system;

[0016] K Gi To determine the capacity share of each unit, the droop coefficient of the aggregation model is defined as follows:

[0017] (1);

[0018] Here, R is the equivalent gain parameter, N is the total number of gas turbine units, and R is the equivalent gain parameter. Gi Let i be the governor gain of each gas turbine unit, and i be the serial number of the gas turbine unit in the system. To simplify the aggregation model, the governor units are normalized, and the standardized gain of each unit branch is defined. for:

[0019] (2);

[0020] in,

[0021] (3);

[0022] Based on the load-side power of the power system, calculate the capacity percentage of each type of generator unit, i.e., the per-unit value of the system, and then compare the rotor inertia time constant of the generator unit with this value K. Gi Multiplying these components yields a simplified aggregate model of the power system for the generator and load components, namely:

[0023] (4);

[0024] Where H is the rotor inertia parameter of the convergent model, Hi These are the rotor inertia parameters for each unit.

[0025] Furthermore, step 2, which involves performing model aggregation on the speed control system of the generator set, includes:

[0026] Let the aggregation parameter Its system parameters are aggregated as follows:

[0027] (5).

[0028] Furthermore, the correctness of the aggregation model is verified, and mathematical error analysis is performed on the errors; in the dual-machine aggregation, the speed controller part is factored, including:

[0029] Factorize the governor model of the gas turbine unit:

[0030] (6);

[0031] Where s is a variable in the system frequency domain; A1 and A2 are process parameters, specifically:

[0032] (7);

[0033] (8);

[0034] Its denominator s 2 coefficient is , and The coefficients are all relatively small, so their product is even smaller. Therefore, s is ignored in the derivation. 2 The impact of the item; and the b of the gas turbine unit g The coefficient is small and has little impact on the overall speed controller control loop. In summary, after derivation, equation (6) can be decomposed as follows:

[0035] (9);

[0036] Among them, B i1 B i2 The coefficients are the process coefficients for factorization.

[0037] Specifically as follows:

[0038] (10);

[0039] (11);

[0040] (12);

[0041] (13);

[0042] (14);

[0043] (15);

[0044] (16);

[0045] Among them, M, B 11 B 12 B 21 B 22 K1 and K2 are both process parameters.

[0046] Furthermore, step 3 involves deriving the error of the speed controller's aggregation model, including: dual-machine aggregation error derivation and multi-machine parameter aggregation error derivation.

[0047] Furthermore, the derivation of the dual-machine aggregation error includes:

[0048] The single-unit speed governor model is divided into two parts for error calculation. The classification is based on the power exponent of the fraction: errors with the same power exponent are categorized as follows: Errors different from those of fractional exponents ;

[0049] (1) Errors of fractions with the same exponent The derivation is as follows:

[0050] (17);

[0051] Among them, T CD1 T is the compression displacement time constant of the first unit. CD2 T is the compression displacement time constant of the second unit. CD Let M1 be the compression displacement time constant of the polymerization unit, M2 be the process parameters of the first and second units respectively, and M be the process parameter of the polymerization unit. , q0, q1, q2, q3 are the normalized gains of the first and second generating units, respectively; p0, p1, p2, p3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the numerator of the error function, respectively; q0, q1, q2, q3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the denominator of the error function, respectively.

[0052] After derivation, the undetermined coefficients for each term are:

[0053] (18);

[0054] (19);

[0055] (20);

[0056] (twenty one);

[0057] (twenty two);

[0058] (twenty three);

[0059] (twenty four);

[0060] (25);

[0061] It can be seen that the coefficient of the cubic term is 0, and the coefficients of its quadratic and linear terms differ significantly, that is... Therefore, we can conclude that:

[0062] (26);

[0063] (2) Errors due to different exponents of the fraction The derivation is as follows:

[0064] related The proof is as follows:

[0065] Will Divided into two parts, namely

[0066] (27);

[0067] in,

[0068] (28);

[0069] (29);

[0070] (30);

[0071] (31);

[0072] (32);

[0073] Through mathematical analysis and derivation, it can be seen that the coefficients of the cubic and linear terms are both 0, while the coefficient of the quadratic term differs significantly: that is... ;

[0074] (33);

[0075] Similarly, it can be proven that:

[0076] (34);

[0077] Right now:

[0078] (35);

[0079] Therefore, we can conclude that: ;

[0080] Furthermore, the derivation of the multi-machine parameter aggregation error is as follows:

[0081] Group them, that is, aggregate them in groups of 2, and then aggregate the aggregated groups, using parameter T. CD The mathematical derivation and proof are as follows:

[0082] definition The parameters are the aggregated parameters of the first t units, i.e.:

[0083] (36);

[0084] Among them, the equivalent gain parameters of the first t units for K G This represents the percentage of power capacity. T is the equivalent gain parameter. CD It is the compression displacement time constant, T CDi It is the compression displacement time constant of the i-th unit;

[0085] Based on the definition of normalized gain, generalizing it to the (t+1)th term yields:

[0086] (37);

[0087] (38);

[0088] in, For the first t standardized gains; according to the above formula, for the parameters of the first t+1 units The derivation is as follows:

[0089] (39);

[0090] The other parameters are consistent with the derivation and proof process of the dual-machine aggregation error.

[0091] Furthermore, the model described in step 4 consists of two parts: a speed governor system and a power generation system.

[0092] Furthermore, the speed governor system includes: a speed governor inertial element, a valve positioner element, a fuel system inertial element, and a mechanical gain element; the power generation system is characterized by: a generator set rotor inertial element.

[0093] A computer storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of implementing the frequency response aggregation model of the weighted order reduction multi-gas turbine system are disclosed.

[0094] The present invention has the following beneficial effects and advantages:

[0095] In the process of high-frequency stability analysis, the solution efficiency of the model needs to be ensured to be both simple and efficient, while also possessing high accuracy, so as to avoid the occurrence of analysis results that are lagging or inaccurate and thus mismatched with the power grid's safety and stability.

[0096] This invention enables a simplified model of the frequency dynamic process. Firstly, based on the concept of "weighted average" aggregation of the frequency response of multiple thermal power unit systems, the governor component is factored to obtain a simplified analytical representation of the governor. Then, aggregation of two-unit gas turbine systems is performed to complete the error derivation and proof of the governor component. Finally, based on the above proof and analysis, the droop coefficient of the aggregated model is obtained, thus yielding a frequency response aggregation model of a multi-gas turbine unit system containing only one set of governor inertial components. This invention simplifies and reduces the order of the frequency response model for multi-gas turbine unit systems by using the weighted average concept, enabling rapid analysis of the system's frequency response process.

[0097] This invention simplifies and analyzes the speed governor model, completing the core of model order reduction. It aggregates parameters and simplifies the model from multiple machines to a single machine, which can further simplify the model. Theoretically, it has certain advantages in solving the dynamic characteristics of the model. Attached Figure Description

[0098] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0099] Figure 1 This is a flowchart of the frequency response aggregation model of the present invention;

[0100] Figure 2 This is a comparison chart of the results of the frequency response model of the multi-unit gas turbine system and the frequency response aggregation model of the multi-gas turbine system of the present invention; Detailed Implementation

[0101] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0102] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0103] The following reference Figures 1-2 The technical solutions of some embodiments of the present invention are described below.

[0104] Example 1

[0105] This invention provides an embodiment of a frequency response aggregation model for a multi-gas turbine system based on weighted order reduction. For example... Figure 1 As shown, Figure 1 This is a flowchart of the frequency response aggregation model of the present invention.

[0106] The present invention specifically includes the following steps:

[0107] Step 1: Factorize the speed controller section.

[0108] Based on the control principle of gas turbine units, a system frequency response model is established, including the governor, valve positioner, fuel system, combustion chamber, and generator.

[0109] Define R G K is the system's droop coefficient. G c represents the power capacity ratio. g and b g X represents the coefficients of the transfer function of the valve positioner in the fuel system. g It is the lead time constant of the speed governor section, Y g It is the lag time constant of the speed governor section, T CR It is the combustion reaction delay time constant, T F It is the fuel system time constant, T CD It is the compression displacement time constant.

[0110] Based on the concept of unit aggregation model, the parameters of the Multi-machine System Frequency Response Model (MM-SFR) are aggregated to form the Aggregated Multi-machine Gas Turbine System Frequency Response Model (ASFR).

[0111] K Gi To determine the capacity share of each unit, the droop coefficient of the aggregation model is defined as follows:

[0112] (1);

[0113] Here, R is the equivalent gain parameter, N is the total number of gas turbine units, and R is the equivalent gain parameter. Gi Let i be the governor gain of each gas turbine unit, and i be the serial number of the gas turbine unit in the system. To simplify the aggregation model, the governor units are normalized, and the standardized gain of each unit branch is defined. for:

[0114] (2);

[0115] in,

[0116] (3);

[0117] In the generator and load sections, each unit has different model inertia times and different system capacities, but their generator structures are basically the same. Therefore, based on the load-side power of the power system, the capacity ratio of each type of unit is calculated, i.e., the per-unit value of the system. The rotor inertia time constant of the generator unit is then compared with this value K. Gi Multiplying these components yields a simplified aggregate model of the power system, specifically the generator and load components:

[0118] (4);

[0119] Where H is the rotor inertia parameter of the convergent model, H i These are the rotor inertia parameters for each unit.

[0120] Step 2: Perform model aggregation in the speed control system section of the generator set:

[0121] Let the aggregation parameter Its system parameters are aggregated as follows:

[0122] (5);

[0123] To verify the correctness of the model, mathematical error analysis is needed. Taking dual-machine aggregation as an example, the speed governor part is factored:

[0124] Factorize the governor model of the gas turbine unit:

[0125] (6);

[0126] Where s is a variable in the system frequency domain; A1 and A2 are process parameters, specifically:

[0127] (7);

[0128] (8);

[0129] Its denominator s 2 coefficient is ,generally and Since the coefficients are all relatively small, their product is even smaller, and s can be ignored during the derivation. 2 The impact of the item. And typically, gas turbine units b g The coefficient is small and has little impact on the overall speed controller control loop. In summary, after derivation, equation (6) can be decomposed as follows:

[0130] (9);

[0131] Among them, B i1 B i2 The coefficients are the process coefficients for factorization.

[0132] Specifically:

[0133] (10);

[0134] (11);

[0135] (12);

[0136] (13);

[0137] (14);

[0138] (15);

[0139] (16);

[0140] Among them, M, B11 B 12 B 21 B 22 K1 and K2 are both process parameters.

[0141] Step 3: Derive the error of the aggregate model of the speed controller.

[0142] The error derivation for the speed governor aggregation stage includes: dual-machine aggregation error derivation and multi-machine parameter aggregation error derivation.

[0143] (1) Derivation of the dual-machine aggregation error.

[0144] Taking dual-machine aggregation as an example, the error derivation of the speed governor stage is performed. As can be seen from Equation 3, the single-machine speed governor model is quite complex, making direct error calculation difficult. Therefore, the error is calculated in two parts, categorized according to the power exponent of the fraction, including errors with the same power exponent. Errors different from those of fractional exponents .

[0145] First, we need to account for errors when the exponents of the fractions are the same. Derivation:

[0146] (17);

[0147] Among them, T CD1 T is the compression displacement time constant of the first unit. CD2 T is the compression displacement time constant of the second unit. CD Let M1 be the compression displacement time constant of the polymerization unit, M2 be the process parameters of the first and second units respectively, and M be the process parameter of the polymerization unit. , q0, q1, q2, q3 are the normalized gains of the first and second generating units, respectively; p0, p1, p2, p3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the numerator of the error function, respectively; q0, q1, q2, q3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the denominator of the error function, respectively.

[0148] After derivation, the undetermined coefficients for each term are:

[0149] (18);

[0150] (19);

[0151] (20);

[0152] (twenty one);

[0153] (twenty two);

[0154] (twenty three);

[0155] (twenty four);

[0156] (25);

[0157] Through mathematical analysis and derivation, it can be seen that the coefficient of the cubic term is 0, and the coefficients of its quadratic and linear terms differ significantly. Therefore, we can conclude that:

[0158] (26);

[0159] Secondly, the error varies depending on the exponent of the fraction. The derivation is as follows:

[0160] related The proof is as follows:

[0161] Will Divided into two parts, namely

[0162] (27);

[0163] in,

[0164] (28);

[0165] (29);

[0166] (30);

[0167] (31);

[0168] (32);

[0169] Through mathematical analysis and derivation, it can be seen that the coefficients of the cubic and linear terms are both 0, while the coefficient of the quadratic term differs significantly: that is... ;

[0170] (33);

[0171] Similarly, it can be proven that:

[0172] (34);

[0173] Right now:

[0174] (35);

[0175] Therefore, it can be concluded that .

[0176] (2) Derivation of multi-machine parameter aggregation error.

[0177] The derivation and proof of the multi-machine parameter aggregation error are similar to those of the two-machine aggregation error. It only requires grouping them, i.e., aggregating two machines as a group, and then aggregating the aggregated groups. Let parameter T... CD We will use an example to provide a mathematical derivation and proof.

[0178] definition The parameters are the aggregated parameters of the first t units, i.e.:

[0179] (36);

[0180] Among them, the equivalent gain parameters of the first t units for K G This represents the percentage of power capacity. T is the equivalent gain parameter. CD It is the compression displacement time constant, T CDi It is the compression displacement time constant of the i-th unit.

[0181] Based on the normalized gain definition in step 1, we generalize it to the (t+1)th term to obtain:

[0182] (37);

[0183] (38);

[0184] in, This represents the standardized gain for the first t units. Based on the above formula, the parameters for the first t+1 units... The following derivation was made:

[0185] (39);

[0186] The other parameters are consistent with the above-described method for deriving and proving the dual-machine aggregation error, and will not be repeated here.

[0187] Step 4: Based on the above three steps, a frequency response aggregation model for a multi-gas turbine system based on weighted order reduction is obtained.

[0188] The model consists of two parts: a governor system and a power generation system.

[0189] The governor system mainly includes: governor inertial element, valve positioner element, fuel system inertial element, and mechanical gain element.

[0190] The power generation system mainly consists of the rotor inertia element of the generator set.

[0191] The model established in this invention can reduce the order of the frequency response model of a multi-gas turbine system, enabling rapid analysis of the system's frequency response process.

[0192] Example 2

[0193] This invention provides an embodiment of a frequency response aggregation model for a multi-gas turbine system based on weighted order reduction. For example... Figure 2 As shown, Figure 2 This is a comparison chart of the frequency response model of the multi-unit gas turbine system and the frequency response aggregation model of the multi-gas turbine system of the present invention.

[0194] The parameters are set as follows during the derivation:

[0195] ;

[0196] Its denominator s 2 coefficient is , can be ignored. Because b g The error is relatively small and has little impact on the overall speed controller control loop. Therefore, after derivation in steps one and two, the error is obtained. .

[0197] After deriving equation (17), the undetermined coefficients for each term are:

[0198] ;

[0199] ;

[0200] ;

[0201] It can be seen that the coefficients of the quadratic and linear terms differ significantly and can be ignored. Therefore, we can conclude that: .

[0202] Will Divided into two parts, the undetermined coefficients of equation (28) are:

[0203] ;

[0204] ;

[0205] It can be seen that the coefficients of their quadratic terms differ significantly and can be ignored. Therefore, we can conclude that: .

[0206] Based on the mathematical derivation, this invention was verified through simulation in MATLAB / Simulink. In the example, three gas turbine units were designed with system capacities of 150MW, 100MW, and 120MW, respectively, and the system disturbance was 18.5MW, or 0.05 pu. Error analysis was performed using the lowest points of the two graphs. Hz, Hz. Therefore: The Hz error is 3.7%, which is within the acceptable error range.

[0207] from Figure 2 The simulation results show that the initial frequency change rate, the lowest frequency point, and the final steady-state frequency are basically consistent with the calculation results of the MM-SFR model. Furthermore, the dynamic frequency change process of the two models is basically the same, indicating that the ASFR model has high calculation accuracy and proving the scientific validity and feasibility of the aggregation model.

[0208] Example 3

[0209] Based on the same inventive concept, this embodiment of the invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the frequency response aggregation model of a multi-gas turbine system based on weighted order reduction as described in embodiment 1 or 2.

[0210] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0211] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0212] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0213] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0214] The preferred method of the present invention has been described above. It should be understood that the present invention is not limited to the specific embodiments described above. Devices and structures not described in detail should be understood as being implemented in a conventional manner in the art. Any person skilled in the art can make many possible variations and modifications to the technical methods of the present invention, or modify them into equivalent methods, without departing from the scope of the present invention. This does not affect the essential content of the present invention. Therefore, any simple modifications, equivalent changes, and alterations made to the above methods based on the technical essence of the present invention, without departing from the content of the present invention, still fall within the protection scope of the present invention.

Claims

1. A frequency response convergence model for a multi-gas turbine system based on weighted order reduction, characterized by: Includes the following steps: Step 1. Factorize the governor section. Based on the control principle of the gas turbine unit, establish a system frequency response model including the governor, valve positioner, fuel system, combustion chamber, and generator; define R... G K is the system's droop coefficient. G c represents the power capacity ratio. g and b g X represents the coefficients of the transfer function of the valve positioner in the fuel system. g It is the lead time constant of the speed governor section, Y g It is the lag time constant of the speed governor section, T CR It is the combustion reaction delay time constant, T F It is the fuel system time constant, T CD It is the compression displacement time constant; based on the concept of unit aggregation model, the parameters of the frequency response model of the multi-gas turbine unit system are aggregated to form a frequency response aggregation model of the multi-gas turbine unit system; K Gi To determine the capacity share of each unit, the droop coefficient of the aggregation model is defined as follows: (1); where, Here, R is the equivalent gain parameter, N is the total number of gas turbine units, and R is the equivalent gain parameter. Gi Let i be the governor gain of each gas turbine unit, and i be the serial number of the gas turbine unit in the system. To simplify the aggregation model, the governor units are normalized, and the standardized gain of each unit branch is defined. for: (2); where, (3); Based on the load-side power of the power system, calculate the capacity ratio of each type of unit, i.e., the per-unit value of the system, and compare the rotor inertia time constant of the generator unit with this value K. Gi Multiplying these components yields a simplified aggregate model of the power system for the generator and load components, namely: (4), where H is the rotor inertia parameter of the convergent model, H i These are the rotor inertia parameters for each unit; Step 2. Perform model aggregation on the speed control system of the generator set, including: Let the aggregation parameter Its system parameters are aggregated as follows: (5); Step 3. Error derivation for the speed governor aggregation stage, including: dual-machine aggregation error derivation and multi-machine parameter aggregation error derivation; the dual-machine aggregation error derivation includes: dividing the single-machine speed governor model into two parts for error calculation, and classifying them according to the power exponent of the fraction: errors with the same power exponent. Errors different from those of fractional exponents ; (1) Errors of fractions with the same exponent The derivation is as follows: (17) ； Among them, T CD1 T is the compression displacement time constant of the first unit. CD2 T is the compression displacement time constant of the second unit. CD Let M1 be the compression displacement time constant of the polymerization unit, M2 be the process parameters of the first and second units respectively, and M be the process parameter of the polymerization unit. , q0, q1, q2, q3 are the normalized gains of the first and second generating units, respectively; p0, p1, p2, p3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the numerator of the error function, respectively; q0, q1, q2, q3 are the coefficients of the 0th, 1st, 2nd, and 3rd terms of the polynomial in the denominator of the error function, respectively. After derivation, the undetermined coefficients for each term are: (18) ； (19) ； (20) ； (21) ； (22) ； (23) ； (24) ； (25) ； It can be seen that the coefficient of the cubic term is 0, and the coefficients of its quadratic and linear terms differ significantly, that is... Therefore, we can conclude that: (26); (2) Errors due to different exponents of the fraction The derivation is as follows: related The proof is as follows: It is divided into two parts, namely: (27); in, (28); (29) ； (30) ； (31) ； (32) ； Through mathematical analysis and derivation, it can be seen that the coefficients of the cubic and linear terms are both 0, while the coefficient of the quadratic term differs significantly: that is... ; (33) ； Similarly, it can be proven that: (34); Right now: (35); Therefore, we can conclude that: ; Step 4. Based on the above 3 steps, a frequency response aggregation model of a multi-gas turbine system based on weighted order reduction is obtained, which consists of two parts: a governor system and a power generation system.

2. The frequency response convergence model for a multi-gas turbine system based on weighted order reduction as described in claim 1, characterized in that: The correctness of the aggregation model is verified, and mathematical error analysis is performed on the errors. In dual-machine aggregation, the governor component is factored, including: Factorize the governor model of the gas turbine unit: （6）； Where s is a variable in the system frequency domain; A1 and A2 are process parameters, specifically: （7）； （8）； Its denominator s 2 coefficient is , and The coefficients are all relatively small, so their product is even smaller. Therefore, s is ignored in the derivation. 2 The impact of the item; and the b of the gas turbine unit g The coefficient is small and has little impact on the overall speed controller control loop. In summary, after derivation, equation (6) can be decomposed as follows: （9）； Among them, B i1 B i2 The coefficients are the process coefficients for factorization. Specifically as follows: (10) （11）； （12）； （13）； （14）； （15）； （16）； Among them, M, B 11 B 12 B 21 B 22 K1 and K2 are both process parameters.

3. The frequency response convergence model for a multi-gas turbine system based on weighted order reduction as described in claim 1, characterized in that: The derivation of the multi-machine parameter aggregation error is as follows: Group them, that is, aggregate them in groups of 2, and then aggregate the aggregated groups, using parameter T. CD The mathematical derivation and proof are as follows: definition The parameters are the aggregated parameters of the first t units, i.e.: （36）； Among them, the equivalent gain parameters of the first t units for K G This represents the percentage of power capacity. T is the equivalent gain parameter. CD It is the compression displacement time constant, T CDi It is the compression displacement time constant of the i-th unit; Based on the definition of normalized gain, generalizing it to the (t+1)th term yields: （37）； （38）； in, For the first t standardized gains; according to the above formula, for the parameters of the first t+1 units The derivation is as follows: （39）； The other parameters are consistent with the derivation and proof process of the dual-machine aggregation error.

4. The frequency response convergence model for a multi-gas turbine system based on weighted order reduction as described in claim 1, characterized in that: The speed governor system includes: a speed governor inertial element, a valve positioner element, a fuel system inertial element, and a mechanical gain element; the power generation system is represented by: a generator set rotor inertial element.

5. A computer storage medium, characterized in that: The computer storage medium contains a computer program, which, when executed by a processor, implements the steps of any one of claims 1-4 regarding the frequency response aggregation model of a multi-gas turbine system based on weighted order reduction.

Images