Gas turbine system predictive control method, device, equipment and medium

Abstract

The invention relates to a gas turbine system predictive control method, device, equipment and medium, and relates to the technical field of gas turbine control, and the method comprises the steps: building a transfer function model of a gas turbine system at a plurality of preset load points through a step response experiment, a prediction controller is designed based on the transfer function models corresponding to the different load points, so that a control increment output by the prediction controller is obtained, and the prediction controller comprises a plurality of sub-prediction controllers corresponding to the different load point models; correcting the control increment which does not accord with a preset constraint condition based on an improved He-Make algorithm; and based on the deviation between the control increment corresponding to each sub-prediction controller and the actual output value, performing weighting processing on the control increment sequence corresponding to each sub-controller by using an improved recursive Bayesian weighting algorithm to obtain a final control increment. According to the method, the operation stability and the response speed of the heavy duty gas turbine system can be remarkably improved, and the strong performance advantage is shown in the aspect of set value tracking.

Description

Technical Field

[0001] This invention relates to the field of gas turbine control technology, and in particular to a predictive control method, device, equipment and medium for a gas turbine system. Background Technology

[0002] Heavy-duty gas turbines are a core component of combined gas-steam systems, playing a crucial role in modern power systems due to their superior energy conversion efficiency. These systems effectively improve power generation efficiency, optimize energy utilization structure, and reduce peak-to-valley differences during grid peak shaving, thereby alleviating power supply shortages. Furthermore, the efficient operation of gas turbine systems enhances grid stability, system security, and economic efficiency. The control system, as a key technology in gas turbine system design, determines the system's performance and safety factor. The nonlinear and time-varying characteristics of gas turbines, along with the complexity of their operating conditions, place higher demands on the control system.

[0003] However, as the demand for operational flexibility in gas turbines continues to increase, traditional PID control methods are gradually showing their limitations under complex dynamic conditions. When the system load fluctuates significantly, traditional control strategies not only struggle to guarantee control accuracy but may also trigger system oscillations or even instability. Therefore, ensuring the flexible and efficient operation of heavy-duty gas turbine systems places higher demands on the design of their control systems. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a predictive control method, device, equipment and medium for a gas turbine system, aiming to solve at least one of the above-mentioned technical problems.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, this application provides a predictive control method for a gas turbine system, employing the following technical solution:

[0007] A predictive control method for a gas turbine system includes:

[0008] Obtain the current operating data of the gas turbine at the current moment, including the current fuel quantity and current output power;

[0009] Based on the current operating data and the preset predictive controller, the control information of the gas turbine at the current moment is calculated. The preset predictive controller includes multiple sub-predictive controllers. Each sub-predictive controller is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by multiple sub-predictive controllers.

[0010] Based on each initial control increment and preset constraints, multiple target control increment sequences are determined;

[0011] Based on the actual control quantity of the gas turbine at the current moment and the sequence of target control increments, the final control increment of the gas turbine at the current moment is determined.

[0012] The beneficial effects of this invention are as follows: By acquiring the current operating data of the gas turbine and combining it with a pre-set predictive controller containing multiple sub-predictive controllers, the initial control increment sequence output by the multiple sub-predictive controllers is calculated. The predictive controller is constructed using historical operating data of different target load points, making the prediction more realistic. The target control increment sequence is determined according to preset constraints to ensure that the control increment meets the system requirements. Then, the final control increment is determined by combining the actual control quantity, which can accurately obtain the control increment suitable for the current moment, effectively addressing the nonlinear characteristics and time-varying nature of the gas turbine load system, improving the stability and safety of unit operation, optimizing control accuracy, and demonstrating strong performance advantages in setpoint tracking.

[0013] Based on the above technical solution, the present invention can be further improved as follows.

[0014] Furthermore, the method for constructing the preset predictive controller includes:

[0015] Based on the actual operating conditions of the gas turbine, multiple target load points of the gas turbine are selected;

[0016] After conducting step response tests on the gas turbine at each target load point, historical operating data of the gas turbine at each target load point is obtained. The historical operating data includes historical fuel consumption and historical output power.

[0017] Based on the historical operating data of each target load point and the preset load system linear method, a transfer function model of the gas turbine is established at each target load point;

[0018] Based on the preset GPC algorithm, the transfer function model of each target load point, and the preset objective function, a sub-predictive controller for the gas turbine at each target load point is constructed. The preset objective function represents a function that measures the expected tracking of the gas turbine output power by the control increment.

[0019] Based on each of the sub-predictive controllers, a predictive controller is constructed.

[0020] The beneficial effects of adopting the above-mentioned further scheme are: it can select target load points according to the actual operating conditions of the gas turbine, obtain historical operating data of each target load point, establish a transfer function model, and design a sub-predictive controller for each target load point based on the transfer function model of the controlled object at different load points and the GPC algorithm, thereby obtaining the control increment of the predictive controller, so as to better adapt to the complex operating conditions of the gas turbine and improve the accuracy and pertinence of the prediction of the control increment of the gas turbine.

[0021] Furthermore, the determination of multiple target control increment sequences based on each initial control increment and preset constraints includes:

[0022] For each initial control increment sequence, determine whether the initial control increment sequence satisfies the preset constraint conditions;

[0023] For each initial control increment sequence, if the initial control increment sequence satisfies a preset constraint condition, then the initial control increment sequence is determined to be the target control increment sequence.

[0024] For each initial control increment sequence, if the initial control increment sequence does not meet the preset constraints, the initial control increment sequence is optimized based on the preset improved hippo optimization algorithm to obtain the target control increment sequence.

[0025] The beneficial effects of adopting the above-mentioned further scheme are: the initial control increment sequence that meets the constraints can be selected as the target control increment sequence. For the initial control increment sequence that does not meet the constraints, the improved Hippo optimization algorithm is used for optimization, which solves the problems of slow convergence speed and inaccurate parameters of the controlled object under the constraints. Finally, a suitable target control increment sequence is obtained, which helps to improve the stability, safety and control accuracy of the gas turbine system control.

[0026] Furthermore, determining whether the initial control increment sequence satisfies preset constraints includes:

[0027] If the current fuel quantity is within a set fuel quantity threshold range, the current output power is within a set output power threshold range, and the initial control increment sequence is within a set control increment threshold, then the initial control increment sequence is determined to satisfy a preset constraint condition.

[0028] The beneficial effects of adopting the above-mentioned further scheme are as follows: After obtaining the current fuel quantity, current output power, and initial control increment sequence output by multiple sub-predictive controllers of the gas turbine, it is possible to determine whether the initial control increment sequence meets the preset constraints based on the set fuel quantity threshold range, output power threshold range, and control increment threshold. This provides a basis for subsequently determining the target control increment sequence, ensuring that the control increment meets the constraints of gas turbine operation, which helps to improve the stability and safety of gas turbine operation and optimize control accuracy.

[0029] Furthermore, the initial control increment sequence is optimized based on a preset improved hippo optimization algorithm to obtain the target control increment sequence, including:

[0030] S31. Based on the set parameter information, the preset Gaussian mapping strategy and the preset Sine mapping strategy, initialize the hippopotamus population. The hippopotamus population includes multiple individuals. The hippopotamus population represents a set of different control increment sequence schemes. Each individual represents a control increment sequence scheme. The set parameter information includes the number of individuals, the maximum number of iterations, the current position of each individual and a random number.

[0031] S32, based on the position of each individual in the hippopotamus population in the current iteration and the preset fitness function, calculate the fitness of each individual in the current iteration, wherein the fitness characterizes the accuracy of the prediction of the control increment sequence scheme corresponding to the individual;

[0032] S33, Based on the fitness of each individual in the current iteration, determine the global best and worst individual positions of the hippopotamus population in the current iteration;

[0033] S34. Based on the current iteration number and the maximum iteration number, determine the adaptive weight factor for the current iteration;

[0034] S35, Based on the globally optimal individual position and adaptive weight factor in the hippopotamus population in the current iteration, calculate the position of each individual in the hippopotamus population during the search phase;

[0035] S36. Based on the position of each individual in the hippopotamus population in the search phase, the position of the globally optimal individual in the hippopotamus population, the position of the worst individual, the dynamic perturbation factor, and the upper and lower boundaries of the solution space in the current iteration, calculate the position of each individual in the hippopotamus population in the development phase. The upper and lower boundaries of the solution space represent the constraints on the control increment sequence scheme corresponding to the individual.

[0036] S37, Based on the position of each individual in the hippopotamus population during the development phase, a new hippopotamus population is generated, and S32 to S37 are executed until the iteration number is satisfied. The control increment sequence scheme corresponding to the individual with the global optimal solution in the hippopotamus population in the current iteration cycle is taken as the target control increment sequence. The current iteration cycle represents the cycle corresponding to iterations S32 to S37.

[0037] The beneficial effects of adopting the above-mentioned further scheme are as follows: initializing the hippopotamus population using a preset Gaussian mapping strategy and a preset Sine mapping strategy can ensure that candidate solutions are evenly distributed in the search space, thereby improving search efficiency; calculating individual fitness can evaluate the predictive accuracy of the control increment sequence scheme; determining the globally optimal and worst individual positions can clarify the search direction; adaptive weight factors can dynamically adjust the search range, enabling individuals to have strong global search capabilities in the early stages of the search and gradually converge in the later stages; calculating the positions of individuals in the search and development stages, and introducing dynamic perturbation factors and considering the upper and lower boundaries of the solution space can optimize the search direction, improve the global search capability of the algorithm, avoid getting trapped in local optima, and ensure the feasibility of the solution; after multiple iterations, the control increment sequence scheme corresponding to the globally optimal solution is output as the target control increment sequence, which can ensure that the optimal control increment that satisfies the constraints is finally obtained.

[0038] Furthermore, the initialization of the hippopotamus population based on the set parameter information, the preset Gaussian mapping strategy, and the preset Sine mapping strategy includes:

[0039] If the random number of an individual is less than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Gaussian mapping strategy and the set parameter information.

[0040] If the random number of an individual is not greater than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Sine mapping strategy and the set parameter information.

[0041] The beneficial effects of adopting the above-mentioned further scheme are: based on the comparison results of individual random numbers and set random number thresholds, selectively using Gaussian mapping strategy or Sine mapping strategy and combining it with set parameter information to initialize the hippo population can ensure that candidate solutions are evenly distributed in the search space, improve the global search capability and search efficiency of the improved hippo optimization algorithm, and thus more effectively optimize the results of incremental prediction of gas turbine system control.

[0042] Furthermore, determining the final control increment of the gas turbine at the current moment based on the actual control quantity of the gas turbine at the previous moment and each of the target control increment sequences includes:

[0043] Based on the actual control quantity of the gas turbine at the current moment and the previous moment and each target control increment sequence, calculate the deviation between the target control increment sequence corresponding to each sub-controller and the actual control quantity.

[0044] Based on each deviation and a preset improved recursive Bayesian weighted algorithm, the target control increment sequence corresponding to each sub-controller is weighted to obtain the final control increment of the gas turbine at the current moment.

[0045] The beneficial effects of adopting the above-mentioned further scheme are as follows: First, calculate the deviation between the target control increment sequence corresponding to each sub-controller and the actual control quantity at the previous moment. Then, use the improved recursive Bayesian weighted algorithm to weight the target control increment sequence. This can fully consider the differences between the output of different sub-controllers and the actual control quantity, thereby accurately determining the final control increment of the gas turbine at the current moment, which helps to improve the accuracy and stability of the gas turbine system control.

[0046] Secondly, this application provides a predictive control device for a gas turbine system, which adopts the following technical solution:

[0047] A predictive control device for a gas turbine system, comprising:

[0048] The acquisition module is used to acquire the current operating data of the gas turbine at the current moment, including the current fuel quantity and the current output power;

[0049] The prediction module is used to calculate the control information of the gas turbine at the current moment, which is output by the prediction controller, based on the current operating data and the preset prediction controller. The preset prediction controller includes multiple sub-predictive controllers, each of which is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by the multiple sub-predictive controllers.

[0050] The constraint module is used to determine multiple target control increment sequences based on each initial control increment and preset constraint conditions;

[0051] The determination module is used to determine the final control increment of the gas turbine at the current moment based on the actual control quantity of the gas turbine at the previous moment and the sequence of each target control increment.

[0052] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0053] An electronic device includes a memory and a processor, wherein the memory stores a computer program capable of being loaded by the processor and executing the predictive control method for a gas turbine system according to any one of the first aspects.

[0054] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0055] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the gas turbine control method according to any one of the first aspects.

[0056] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating a gas turbine control incremental prediction method according to an embodiment of the present invention.

[0058] Figure 2 This is a schematic diagram of the improved hippo optimization algorithm provided in one embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram of the structure of a gas turbine control incremental prediction device according to an embodiment of the present invention;

[0060] Figure 4 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0063] This application provides a predictive control method for a gas turbine system. This method can be executed by an electronic device, which can be a server or a mobile terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The mobile terminal device can be a laptop computer, a desktop computer, etc., but is not limited to these.

[0064] like Figure 1 As shown, a predictive control method for a gas turbine system mainly includes:

[0065] S1, acquire the current operating data of the gas turbine at the current moment, the current operating data including the current fuel quantity and the current output power;

[0066] S2, based on the current operating data and the preset predictive controller, calculate the control information of the gas turbine at the current moment output by the predictive controller. The preset predictive controller includes multiple sub-predictive controllers. Each sub-predictive controller is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by multiple sub-predictive controllers.

[0067] S3, based on each initial control increment and preset constraints, determine multiple target control increment sequences;

[0068] S4. Based on the actual control quantity of the gas turbine at the previous time and the sequence of each target control increment, determine the final control increment of the gas turbine at the current time.

[0069] This method utilizes the current operating data of the gas turbine and combines it with a pre-set predictive controller containing multiple sub-predictive controllers to calculate the initial control increment sequence output by these sub-predictive controllers. The predictive controllers are constructed using historical operating data from different target load points, making the predictions more realistic. The target control increment sequence is determined based on pre-set constraints to ensure that the control increment meets system requirements. Finally, the final control increment is determined by combining the actual control quantity. This method can accurately obtain the control increment suitable for the current moment, effectively addressing the nonlinear characteristics and time-varying nature of the gas turbine load system, improving the stability and safety of unit operation, optimizing control accuracy, and demonstrating strong performance advantages in setpoint tracking.

[0070] Optional, preset methods for constructing predictive controllers include:

[0071] S21, Based on the actual operating conditions of the gas turbine, select multiple target load points for the gas turbine;

[0072] S22, after conducting step response tests on each target load point of the gas turbine, historical operating data of the gas turbine at each target load point is obtained, including historical fuel quantity and historical output power.

[0073] S23, Based on the historical operating data of each target load point and the preset load system linear method, establish the transfer function model of the gas turbine at each target load point;

[0074] S24, based on the preset GPC algorithm, the transfer function model of each target load point and the preset objective function, construct the sub-predictive controller of the gas turbine at each target load point, wherein the preset objective function characterizes the function that measures the control increment to track the expected output power of the gas turbine;

[0075] S25, construct a prediction controller based on each of the sub-prediction controllers.

[0076] In this embodiment, representative load points, such as 100%, 75%, and 50% load, are selected based on the actual operating conditions of the gas turbine. These points cover the key states within the gas turbine's operating range. Then, the gas turbine load system is linearized near each typical operating point to obtain a local linear dynamic model for that operating point. These models accurately reflect the dynamic characteristics of the system near the corresponding load. Subsequently, a sub-predictive controller for each target load point is constructed using the GPC algorithm, the transfer function model for each target load point, and a preset objective function.

[0077] The optimization objective is to minimize the changes in output error and control increment, while ensuring the constraints on input, output, and control increment during the operation of the heavy-duty gas turbine. The objective function is as follows:

[0078]

[0079] In the formula, N is the prediction length, M is the control length, λ is the control weighting constant, y(k+j) is the historical output power of the gas turbine system at time (k+j), ω(k+j) is the expected output power at time (k+j), and ΔU is the control increment.

[0080] Optionally, based on each initial control increment and preset constraints, a multiple target control increment sequence is determined, including:

[0081] For each initial control increment sequence, determine whether the initial control increment sequence satisfies the preset constraint conditions;

[0082] For each initial control increment sequence, if the initial control increment sequence satisfies a preset constraint condition, then the initial control increment sequence is determined to be the target control increment sequence.

[0083] For each initial control increment sequence, if the initial control increment sequence does not meet the preset constraints, the initial control increment sequence is optimized based on the preset improved hippo optimization algorithm to obtain the target control increment sequence.

[0084] In this embodiment of the application, if the current fuel quantity is within a set fuel quantity threshold range, the current output power is within a set output power threshold range, and the initial control increment sequence is within a set control increment threshold, then the initial control increment sequence is determined to satisfy a preset constraint condition.

[0085] The specific constraint formula is as follows:

[0086] u min ≤u(k)≤u max ;

[0087] y min ≤y(k)≤y max ;

[0088] Δu min ≤ΔU≤Δu max ;

[0089] In the formula, umax min Controls the upper and lower limits of the input; ymax min The upper and lower limits of the system output; ΔU is the control increment, i.e., ΔU = u(k) - u(k-1), representing the amplitude of the change in the control signal between adjacent time points; Δumax min Control the upper and lower limits of the increment.

[0090] In the embodiments of this application, such as Figure 2 As shown, the initial control increment sequence is optimized based on a preset improved hippo optimization algorithm to obtain the target control increment sequence, including:

[0091] S31. Based on the set parameter information, the preset Gaussian mapping strategy and the preset Sine mapping strategy, initialize the hippopotamus population. The hippopotamus population includes multiple individuals. The hippopotamus population represents a set of different control increment sequence schemes. Each individual represents a control increment sequence scheme. The set parameter information includes the number of individuals, the maximum number of iterations, the current position of each individual and a random number.

[0092] In this embodiment of the application, the initialization of the hippopotamus population based on the set parameter information, the preset Gaussian mapping strategy, and the preset Sine mapping strategy includes:

[0093] If the random number of an individual is less than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Gaussian mapping strategy and the set parameter information.

[0094] If the random number of an individual is not greater than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Sine mapping strategy and the set parameter information.

[0095] Specifically, the position of an individual in the population is calculated during the initialization phase as follows:

[0096]

[0097] In the formula, X i X represents the current position of individual i; i+1 Let `i` be the updated position of individual `i` in the next generation; `rand` is a random number in the interval [0,1] used to determine the mapping strategy adopted by the current individual; `mod(X)`... i 1) Modulo operation is used to ensure that the individual positions are within a reasonable range and to avoid overflow; 'a' is the chaos factor of the Sine mapping, which is usually in the range of 0.7-1.2 to ensure the uniformity of the mapping values ​​and prevent the solution space search from becoming overly concentrated; sin(π·X) i ) is a Sine mapping function used to introduce chaotic properties and improve the diversity of the search.

[0098] S32, based on the position of each individual in the hippopotamus population in the current iteration and the preset fitness function, calculate the fitness of each individual in the current iteration, wherein the fitness characterizes the accuracy of the prediction of the control increment sequence scheme corresponding to the individual;

[0099] In this embodiment, the preset fitness function is a preset objective function.

[0100] S33, Based on the fitness of each individual in the current iteration, determine the global best and worst individual positions of the hippopotamus population in the current iteration;

[0101] S34. Based on the current iteration number and the maximum iteration number, determine the adaptive weight factor for the current iteration;

[0102] S35, Based on the globally optimal individual position and adaptive weight factor in the hippopotamus population in the current iteration, calculate the position of each individual in the hippopotamus population during the search phase;

[0103] In this embodiment, the hippopotamus individual's position in the river or pond is dynamically updated. An adaptive weighting strategy is introduced to dynamically adjust the search range, giving the individual greater exploration ability in the early stages and gradually allowing it to converge on the optimal solution in the later stages. The formula for updating the individual's position during the search phase is:

[0104] X P =X i ·ω(t)+r1(X best,j -X i )+r2(I2X i -I1X i );

[0105]

[0106] In the formula, X P X represents the position of individual i in the next generation; i Let ω(t) be the current position of individual i; ω(t) is an adaptive weighting factor used to control the search step size, giving the individual greater exploration ability in the early stages and gradually converging in the later stages; T max X represents the maximum number of iterations. best,j I1 and I2 represent the position of the dominant individual; I1 and I2 are random influence factors; r1 and r2 are perturbation factors, with values ​​ranging from [0,1], used to control the degree of individual response to the optimal solution and random factors.

[0107] After an individual's position is updated, if it exceeds the boundary of the search space, it is adjusted to the boundary value to ensure that the search process is carried out within the feasible solution space. The boundary of the search space is a constraint on the control increment sequence scheme corresponding to the individual.

[0108] S36. Based on the position of each individual in the hippopotamus population in the search phase, the position of the globally optimal individual in the hippopotamus population, the position of the worst individual, the dynamic perturbation factor, and the upper and lower boundaries of the solution space in the current iteration, calculate the position of each individual in the hippopotamus population in the development phase. The upper and lower boundaries of the solution space represent the constraints on the control increment sequence scheme corresponding to the individual.

[0109] In this embodiment, to enhance the diversity of solutions and suppress the convergence trend of individual solutions to the boundary during the back-learning process, a back-learning strategy with a dynamic perturbation factor is introduced to further optimize search performance. The improved individual position update formula is as follows:

[0110] X ij =r s ×(l j -u j )-X i,j +λ×(X best,j -X worst,j );

[0111] In the formula, λ is the dynamic perturbation factor, which takes values ​​in the range [-1, 1], and is used to enhance the randomness of the search; X best,j and X worst,j represents the positions of the best and worst individuals in the j-th dimension of the population, respectively; r is a random factor used to increase the search randomness of individuals in the population, thereby improving the diversity of optimization solutions; l j ,u j To solve for the upper and lower boundaries of the space.

[0112] S37, Based on the position of each individual in the hippopotamus population during the development phase, a new hippopotamus population is generated, and S32 to S37 are executed until the iteration number is satisfied. The control increment sequence scheme corresponding to the individual with the global optimal solution in the hippopotamus population in the current iteration cycle is taken as the target control increment sequence. The current iteration cycle represents the cycle corresponding to iterations S32 to S37.

[0113] Initializing the hippopotamus population using pre-defined Gaussian and Sine mapping strategies ensures a uniform distribution of candidate solutions within the search space, improving search efficiency. Calculating individual fitness assesses the predictive accuracy of the control increment sequence scheme. Determining the globally optimal and worst-case individual positions clarifies the search direction. Adaptive weighting factors dynamically adjust the search range, enabling individuals to possess strong global search capabilities in the early stages and gradually converge later. Calculating individual positions during the search and development phases, and introducing dynamic perturbation factors and considering the upper and lower boundaries of the solution space, optimizes the search direction, enhances the algorithm's global search capability, avoids getting trapped in local optima, and ensures solution feasibility. After multiple iterations, the control increment sequence scheme corresponding to the globally optimal solution is output as the target control increment sequence, ensuring that the optimal control increment satisfying the constraints is ultimately obtained.

[0114] Optionally, based on the actual control quantity of the gas turbine at the previous time and the sequence of each target control increment, the final control increment of the gas turbine at the current time is determined, including:

[0115] Based on the actual control quantity of the gas turbine at the current moment and the previous moment and each target control increment sequence, calculate the deviation between the target control increment sequence corresponding to each sub-controller and the actual control quantity.

[0116] Based on each deviation and a preset improved recursive Bayesian weighted algorithm, the target control increment sequence corresponding to each sub-controller is weighted to obtain the final control increment of the gas turbine at the current moment.

[0117] In this embodiment, multi-model predictive control is employed. Based on the deviation between the output of each sub-model and the actual output, an improved recursive Bayesian weighted algorithm is used to weight the outputs of each sub-controller to obtain the final control increment, thus yielding the final control quantity. The first element of this final control quantity is then added to the control system. A new optimal control increment is calculated at the next time step, achieving rolling optimization of the predictive controller. This approach fully considers the differences between the outputs of different sub-controllers and the actual control quantity, thereby accurately determining the final control increment of the gas turbine at the current time step, which helps improve the accuracy and stability of the gas turbine system control.

[0118] Figure 3 A schematic diagram of a predictive control device 200 for a gas turbine system is shown.

[0119] like Figure 3 As shown, a predictive control device 200 for a gas turbine system mainly includes:

[0120] The acquisition module 201 is used to acquire the current operating data of the gas turbine at the current moment, the current operating data including the current fuel quantity and the current output power;

[0121] The prediction module 202 is used to calculate the control information of the gas turbine at the current moment, which is output by the prediction controller, based on the current operating data and the preset prediction controller. The preset prediction controller includes multiple sub-predictive controllers, each of which is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by the multiple sub-predictive controllers.

[0122] The constraint module 203 is used to determine a sequence of multiple target control increments based on each initial control increment and preset constraint conditions;

[0123] The determining module 204 is used to determine the final control increment of the gas turbine at the current moment based on the actual control quantity of the gas turbine at the previous moment and the sequence of each target control increment.

[0124] In one example, the module in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0125] For example, when modules in a device can be implemented via a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these modules can be integrated together as a system-on-a-chip (SOC).

[0126] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.

[0127] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0129] Figure 4 This is a structural block diagram of an electronic device 300 according to an embodiment of this application.

[0130] like Figure 4As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0131] The processor 301 controls the overall operation of the electronic device 300 to complete all or part of the steps in the aforementioned predictive control method for the gas turbine system. The memory 302 stores various types of data to support the operation of the electronic device 300. This data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0132] I / O interface 303 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 304 is used to test wired or wireless communication between electronic device 300 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 304 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0133] The communication bus 305 may include a path for transmitting information between the aforementioned components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 can be divided into an address bus, a data bus, a control bus, etc.

[0134] The electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the gas turbine system predictive control method given in the above embodiments.

[0135] The following describes the computer-readable storage medium provided in the embodiments of this application. The computer-readable storage medium described below can be referred to in correspondence with the gas turbine system predictive control method described above.

[0136] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described predictive control method for a gas turbine system.

[0137] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0138] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0139] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A predictive control method for a gas turbine system, characterized in that, include: Obtain the current operating data of the gas turbine at the current moment, including the current fuel quantity and current output power; Based on the current operating data and the preset predictive controller, the control information of the gas turbine at the current moment is calculated. The preset predictive controller includes multiple sub-predictive controllers. Each sub-predictive controller is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by multiple sub-predictive controllers. Based on each initial control increment and preset constraints, multiple target control increment sequences are determined; Based on the actual control quantity of the gas turbine at the current moment and the sequence of target control increments, the final control increment of the gas turbine at the current moment is determined.

2. The predictive control method for a gas turbine system according to claim 1, characterized in that, The method for constructing the preset predictive controller includes: Based on the actual operating conditions of the gas turbine, multiple target load points of the gas turbine are selected; After conducting step response tests on the gas turbine at each target load point, historical operating data of the gas turbine at each target load point is obtained. The historical operating data includes historical fuel consumption and historical output power. Based on the historical operating data of each target load point and the preset load system linear method, a transfer function model of the gas turbine is established at each target load point; Based on the preset GPC algorithm, the transfer function model of each target load point, and the preset objective function, a sub-predictive controller for the gas turbine at each target load point is constructed. The preset objective function represents a function that measures the expected tracking of the gas turbine output power by the control increment. Based on each of the sub-predictive controllers, a predictive controller is constructed.

3. The predictive control method for a gas turbine system according to claim 1, characterized in that, The determination of multiple target control increment sequences based on each initial control increment and preset constraints includes: For each initial control increment sequence, determine whether the initial control increment sequence satisfies the preset constraint conditions; For each initial control increment sequence, if the initial control increment sequence satisfies a preset constraint condition, then the initial control increment sequence is determined to be the target control increment sequence. For each initial control increment sequence, if the initial control increment sequence does not meet the preset constraints, the initial control increment sequence is optimized based on the preset improved hippo optimization algorithm to obtain the target control increment sequence.

4. The predictive control method for a gas turbine system according to claim 3, characterized in that, The step of determining whether the initial control increment sequence satisfies the preset constraint conditions includes: If the current fuel quantity is within a set fuel quantity threshold range, the current output power is within a set output power threshold range, and the initial control increment sequence is within a set control increment threshold, then the initial control increment sequence is determined to satisfy a preset constraint condition.

5. The predictive control method for a gas turbine system according to claim 3, characterized in that, The initial control increment sequence is optimized using a preset improved hippo optimization algorithm to obtain the target control increment sequence, including: S31. Based on the set parameter information, the preset Gaussian mapping strategy and the preset Sine mapping strategy, initialize the hippopotamus population. The hippopotamus population includes multiple individuals. The hippopotamus population represents a set of different control increment sequence schemes. Each individual represents a control increment sequence scheme. The set parameter information includes the number of individuals, the maximum number of iterations, the current position of each individual and a random number. S32, based on the position of each individual in the hippopotamus population in the current iteration and the preset fitness function, calculate the fitness of each individual in the current iteration, wherein the fitness characterizes the accuracy of the prediction of the control increment sequence scheme corresponding to the individual; S33, Based on the fitness of each individual in the current iteration, determine the global best and worst individual positions of the hippopotamus population in the current iteration; S34. Based on the current iteration number and the maximum iteration number, determine the adaptive weight factor for the current iteration; S35, Based on the globally optimal individual position and adaptive weight factor in the hippopotamus population in the current iteration, calculate the position of each individual in the hippopotamus population during the search phase; S36. Based on the position of each individual in the hippopotamus population in the search phase, the position of the globally optimal individual in the hippopotamus population, the position of the worst individual, the dynamic perturbation factor, and the upper and lower boundaries of the solution space in the current iteration, calculate the position of each individual in the hippopotamus population in the development phase. The upper and lower boundaries of the solution space represent the constraints on the control increment sequence scheme corresponding to the individual. S37, Based on the position of each individual in the hippopotamus population during the development phase, a new hippopotamus population is generated, and S32 to S37 are executed until the iteration number is satisfied. The control increment sequence scheme corresponding to the individual with the global optimal solution in the hippopotamus population in the current iteration cycle is taken as the target control increment sequence. The current iteration cycle represents the cycle corresponding to iterations S32 to S37.

6. The predictive control method for a gas turbine system according to claim 5, wherein initializing the hippopotamus population based on set parameter information, a preset Gaussian mapping strategy, and a preset Sine mapping strategy includes: If the random number of an individual is less than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Gaussian mapping strategy and the set parameter information. If the random number of an individual is not greater than the set random number threshold, the hippopotamus population for the current iteration is generated based on the preset Sine mapping strategy and the set parameter information.

7. The predictive control method for a gas turbine system according to claim 1, characterized in that, The step of determining the final control increment of the gas turbine at the current moment based on the actual control quantity of the gas turbine at the previous moment and each of the target control increment sequences includes: Based on the actual control quantity of the gas turbine at the current moment and the previous moment and each target control increment sequence, calculate the deviation between the target control increment sequence corresponding to each sub-controller and the actual control quantity. Based on each deviation and a preset improved recursive Bayesian weighted algorithm, the target control increment sequence corresponding to each sub-controller is weighted to obtain the final control increment of the gas turbine at the current moment.

8. A predictive control device for a gas turbine system, characterized in that, include: The acquisition module is used to acquire the current operating data of the gas turbine at the current moment, including the current fuel quantity and the current output power; The prediction module is used to calculate the control information of the gas turbine at the current moment, which is output by the prediction controller, based on the current operating data and the preset prediction controller. The preset prediction controller includes multiple sub-predictive controllers, each of which is constructed based on the historical operating data of the gas turbine at different target load points. The target load points represent the load of the gas turbine under different operating conditions. The control information includes the initial control increment sequence output by the multiple sub-predictive controllers. The constraint module is used to determine multiple target control increment sequences based on each initial control increment and preset constraint conditions; The determination module is used to determine the final control increment of the gas turbine at the current moment based on the actual control quantity of the gas turbine at the previous moment and the sequence of each target control increment.

9. An electronic device, characterized in that, Includes a processor, which is coupled to a memory; The processor is configured to execute a computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1-7.

Images