Control method and control device for large rotating machinery

Abstract

The invention provides a control method and device for a large rotating machine, and the method comprises the steps: obtaining the corresponding disturbance state and operation data of the large rotating machine at a current moment in the operation process of the large rotating machine; wherein the disturbance state comprises an external disturbance state and a non-external disturbance state; judging whether accumulative operation data corresponding to the acquired operation data reaches a preset window amount or not, and determining a target fuzzy control model corresponding to the disturbance state based on a judgment result; inputting the operation data into a target fuzzy control model corresponding to the disturbance state, and performing reasoning calculation based on the operation data by using the target fuzzy control model to obtain a target mechanical frequency output by the target fuzzy control model; and controlling a motor arranged in the large rotary machine to operate according to the target mechanical frequency. By means of the method, the stability and safety of operation of the large rotating machine are improved.

Description

Technical Field

[0001] This application relates to the technical field of mechanical industry control, and particularly to a control method and a control device for large rotating machinery. Background Art

[0002] In the industrial field, large rotating machinery is affected by external disturbances and is prone to losing its stable state, thereby affecting the normal operation performance and working efficiency of large rotating machinery, and even triggering operation accidents. However, the existing control methods lack sufficient robustness and fault tolerance for large rotating machinery in a complex disturbance environment.

[0003] Currently, the control methods for large rotating machinery usually adopt methods such as PID control and adaptive control. However, when these methods are used to control large rotating machinery in a complex disturbance environment, problems such as poor anti-interference performance and difficulty in coping with rapidly changing disturbances will occur, reducing the stability of the operation of large rotating machinery, and further reducing the safety of large rotating machinery. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a control method and a control device for large rotating machinery. By online parameter updating and robustness evaluation of the fuzzy control model under different disturbance states, the target fuzzy control model under different disturbance states is obtained, and the operation data of the large rotating machinery is input into the target fuzzy control model under the current disturbance state, and the large rotating machinery is controlled to operate according to the target mechanical frequency output by the target fuzzy control model, so that the large rotating machinery can better cope with complex external environmental changes, improve the anti-interference ability in the external disturbance environment, improve the stability of the operation of the large rotating machinery, and further improve the safety of the large rotating machinery.

[0005] The embodiment of this application provides a control method for large rotating machinery. The control method includes: During the operation of the large rotating machinery, obtain the disturbance state and operation data corresponding to the large rotating machinery at the current moment; wherein, the disturbance state includes an external disturbance state and a state without external disturbance; Judge whether the cumulative operation data corresponding to the obtained operation data reaches a preset window quantity, and based on the judgment result, determine the target fuzzy control model corresponding to the disturbance state; Input the operation data into the target fuzzy control model corresponding to the disturbance state, and use the target fuzzy control model to perform inference calculation based on the operation data to obtain the target mechanical frequency output by the target fuzzy control model; Control the motor arranged in the large rotating machinery to operate according to the target mechanical frequency.

[0006] Further, when the cumulative operation data corresponding to the obtained operation data does not reach the preset window quantity in the judgment result, determining the target fuzzy control model corresponding to the disturbance state based on the judgment result includes: Obtain the fuzzy control model corresponding to the disturbance state updated online last time, and determine the fuzzy control model as the target fuzzy control model corresponding to the disturbance state.

[0007] Further, when the cumulative operation data corresponding to the obtained operation data reaches the preset window quantity in the judgment result, determining the target fuzzy control model corresponding to the disturbance state based on the judgment result includes: Obtain the fuzzy control model corresponding to the disturbance state updated online last time, and use the cumulative operation data to update the model parameters in the fuzzy control model online to obtain the updated fuzzy control model corresponding to the disturbance state; Use the target operation data corresponding to when the updated fuzzy control model is obtained to perform a robustness evaluation on the updated fuzzy control model, obtain a robustness evaluation result, and based on the robustness evaluation result, adjust the updated fuzzy control model corresponding to each disturbance state to obtain the target fuzzy control model corresponding to the disturbance state.

[0008] Further, when the disturbance state is a state of no external disturbance, using the obtained cumulative operation data to update the model parameters in the fuzzy control model online to obtain the updated fuzzy control model corresponding to the disturbance state includes: Input the first cumulative operation data in the cumulative operation data into the first fuzzy control model corresponding to the state of no external disturbance to obtain a first mechanical frequency value output by the first fuzzy control model; Based on the first mechanical frequency value and the first preset mechanical frequency value, use the first objective function preset corresponding to the state of no external disturbance to calculate and determine the first target value corresponding to the first cumulative operation data in the state of no external disturbance; Judge whether the first target value is less than the preset target value; If not, based on the first target value, perform an update calculation on the model parameters in the first fuzzy control model to obtain the first model parameters corresponding to the first update period, and replace the model parameters in the first fuzzy control model with the first model parameters to obtain the first fuzzy control model corresponding to the first update period; Repeatedly input the cumulative operation data into the first fuzzy control model corresponding to each update period, calculate the first target value corresponding to each update period, and when the first target value is greater than or equal to the preset target value, update the model parameters in the first fuzzy control model to obtain the first model parameter corresponding to each update period until the first target value is less than the preset target value; When the first target value is less than the preset target value, determine the first target model parameter corresponding to the current update period, and replace the model parameters in the first fuzzy control model with the first target model parameter to obtain the first updated fuzzy control model corresponding to the state without external disturbance.

[0009] Further, when the disturbance state is an external disturbance state, the method of using the operation data to online update the model parameters in the fuzzy control model to obtain the updated fuzzy control model corresponding to the disturbance state includes: Input the first cumulative operation data in the cumulative operation data into the second fuzzy control model corresponding to the external disturbance state to obtain the second mechanical frequency value output by the second fuzzy control model; Based on the second mechanical frequency value, the second preset mechanical frequency value, the first target model parameter, and the model parameters in the second fuzzy control model, use the second objective function preset corresponding to the external disturbance state to calculate and determine the second target value corresponding to the first cumulative operation data under the external disturbance state; Judge whether the second target value is less than the preset target value; If not, based on the second target value, perform an update calculation on the model parameters in the second fuzzy control model to obtain the second model parameter corresponding to the first update period, and replace the model parameters in the second fuzzy control model with the second model parameter to obtain the second fuzzy control model corresponding to the first update period; Repeatedly input the cumulative operation data into the second fuzzy control model corresponding to each update period, calculate the second target value corresponding to each update period, and when the second target value is greater than or equal to the preset target value, update the model parameters in the second fuzzy control model to obtain the second model parameter corresponding to each update period until the second target value is less than the preset target value; When the second target value is less than the preset target value, determine the second target model parameter corresponding to the current update period, and replace the model parameters in the second fuzzy control model with the second target model parameter to obtain the second updated fuzzy control model corresponding to the external disturbance state.

[0010] Further, the method for performing a robustness evaluation on the updated fuzzy control model by using the target operation data corresponding to when the updated fuzzy control model is obtained to obtain a robustness evaluation result includes: Inputting the target operation data corresponding to when the updated fuzzy control model is obtained into the first updated fuzzy control model corresponding to the state without external disturbance and the second updated fuzzy control model corresponding to the state with external disturbance respectively, to obtain a first evaluated mechanical frequency value output by the first updated fuzzy control model and a second evaluated mechanical frequency value output by the second updated fuzzy control model respectively; Based on the first target model parameter in the first updated fuzzy control model, the second target model parameter in the second updated fuzzy control model, the first evaluated mechanical frequency value, and the second evaluated mechanical frequency value, calculating and determining an objective function value by using a preset objective function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, the first target model parameter, the second target model parameter, and the objective function value, calculating and determining a response characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model by using a preset response evaluation function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, and the target operation data, respectively determining a first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value and a second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value by using the first updated fuzzy control model and the second updated fuzzy control model; Based on the first mechanical frequency change rate, the second mechanical frequency change rate, the first target model parameter, the second target model parameter, and the objective function value, calculating and determining a fluctuation characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model by using a preset fluctuation evaluation function.

[0011] Further, the method for inputting the operation data into the target fuzzy control model corresponding to the disturbance state, and performing inference calculation on the operation data by using the target fuzzy control model to obtain a target mechanical frequency output by the target fuzzy control model includes: Inputting the operation data into an input layer set in the target fuzzy control model corresponding to the disturbance state to obtain an input vector output by the input layer; Inputting the input vector into a radial basis layer set in the target fuzzy control model, and performing fuzzy processing on the input vector by using the radial basis layer to obtain a radial basis output vector output by the radial basis layer; Input the radial basis output vector into the normalization layer set in the target fuzzy control model, and use the normalization layer to perform normalization processing on the radial basis output vector to obtain the normalized output vector output by the normalization layer; Input the normalized output vector into the output layer set in the target fuzzy control model to obtain the mechanical frequency change amount output by the output layer; Obtain the historical target mechanical frequency corresponding to the large rotating machine at the previous moment, and determine the sum of the historical target mechanical frequency and the mechanical frequency change amount as the target mechanical frequency output by the target fuzzy control model.

[0012] An embodiment of the present application further provides a control device for a large rotating machine, and the control device includes: A data acquisition module, configured to acquire the disturbance state and operation data corresponding to the large rotating machine at the current moment during the operation of the large rotating machine; wherein, the disturbance state includes an external disturbance state and a state without external disturbance; A parameter update module, configured to determine whether the cumulative operation data corresponding to the acquired operation data reaches a preset window quantity, and based on the judgment result, determine the target fuzzy control model corresponding to the disturbance state; A model inference module, configured to input the operation data into the target fuzzy control model corresponding to the disturbance state, and use the target fuzzy control model to perform inference calculation based on the operation data to obtain the target mechanical frequency output by the target fuzzy control model; A mechanical control module, configured to control the motor arranged in the large rotating machine to operate according to the target mechanical frequency.

[0013] Further, when the judgment result is that the cumulative operation data corresponding to the acquired operation data does not reach the preset window quantity, when the parameter update module is used to determine the target fuzzy control model corresponding to the disturbance state based on the judgment result, the parameter update module is used for: Obtain the fuzzy control model corresponding to the disturbance state updated online last time, and determine the fuzzy control model as the target fuzzy control model corresponding to the disturbance state.

[0014] Further, when the judgment result is that the cumulative operation data corresponding to the acquired operation data reaches the preset window quantity, when the parameter update module is used to determine the target fuzzy control model corresponding to the disturbance state based on the judgment result, the parameter update module is used for: Obtain the fuzzy control model corresponding to the disturbance state of the previous online update, and use the cumulative operation data to online update the model parameters in the fuzzy control model to obtain the updated fuzzy control model corresponding to the disturbance state; Use the target operation data corresponding to the obtained updated fuzzy control model to perform a robustness evaluation on the updated fuzzy control model to obtain a robustness evaluation result, and based on the robustness evaluation result, adjust the updated fuzzy control model corresponding to each disturbance state to obtain the target fuzzy control model corresponding to the disturbance state.

[0015] Further, when the disturbance state is a state of no external disturbance, when the parameter update module is used to online update the model parameters in the fuzzy control model by using the obtained cumulative operation data to obtain the updated fuzzy control model corresponding to the disturbance state, the parameter update module is used for: Input the first cumulative operation data in the cumulative operation data into the first fuzzy control model corresponding to the state of no external disturbance to obtain a first mechanical frequency value output by the first fuzzy control model; Based on the first mechanical frequency value and the first preset mechanical frequency value, use the first target function preset corresponding to the state of no external disturbance to calculate and determine the first target value corresponding to the first cumulative operation data in the state of no external disturbance; Judge whether the first target value is less than the preset target value; If not, then based on the first target value, perform an update calculation on the model parameters in the first fuzzy control model to obtain the first model parameters corresponding to the first update period, and replace the model parameters in the first fuzzy control model with the first model parameters to obtain the first fuzzy control model corresponding to the first update period; Repeat inputting the cumulative operation data into the first fuzzy control model corresponding to each update period, calculate the first target value corresponding to each update period, and when the first target value is greater than or equal to the preset target value, update the model parameters in the first fuzzy control model to obtain the first model parameters corresponding to each update period until the first target value is less than the preset target value; When the first target value is less than the preset target value, determine the first target model parameters corresponding to the current update period, and replace the model parameters in the first fuzzy control model with the first target model parameters to obtain the first updated fuzzy control model corresponding to the state of no external disturbance.

[0016] Further, when the perturbation state is an external perturbation state, when the parameter update module is used to online update the model parameters in the fuzzy control model by using the operation data to obtain the updated fuzzy control model corresponding to the perturbation state, the parameter update module is used for: Input the first cumulative operation data in the cumulative operation data into the second fuzzy control model corresponding to the external perturbation state to obtain a second mechanical frequency value output by the second fuzzy control model; Based on the second mechanical frequency value, the second preset mechanical frequency value, the first target model parameter, and the model parameters in the second fuzzy control model, use the second target function preset corresponding to the external perturbation state to calculate and determine the second target value corresponding to the first cumulative operation data under the external perturbation state; Judge whether the second target value is less than the preset target value; If not, based on the second target value, perform an update calculation on the model parameters in the second fuzzy control model to obtain the second model parameters corresponding to the first update period, and replace the model parameters in the second fuzzy control model with the second model parameters to obtain the second fuzzy control model corresponding to the first update period; Repeat inputting the cumulative operation data into the second fuzzy control model corresponding to each update period, calculate the second target value corresponding to each update period, and when the second target value is greater than or equal to the preset target value, update the model parameters in the second fuzzy control model to obtain the second model parameters corresponding to each update period until the second target value is less than the preset target value; When the second target value is less than the preset target value, determine the second target model parameters corresponding to the current update period, and replace the model parameters in the second fuzzy control model with the second target model parameters to obtain the second updated fuzzy control model corresponding to the external perturbation state.

[0017] Further, when the parameter update module is used to perform a robustness evaluation on the updated fuzzy control model by using the target operation data corresponding to obtaining the updated fuzzy control model to obtain a robustness evaluation result, the parameter update module is used for: Input the target operation data corresponding to obtaining the updated fuzzy control model into the first updated fuzzy control model corresponding to the no-external-perturbation state and the second updated fuzzy control model corresponding to the external perturbation state respectively, and obtain a first evaluation mechanical frequency value output by the first updated fuzzy control model and a second evaluation mechanical frequency value output by the second updated fuzzy control model respectively; Based on the first target model parameters in the first updated fuzzy control model, the second target model parameters in the second updated fuzzy control model, the first evaluated mechanical frequency value, and the second evaluated mechanical frequency value, calculate and determine the objective function value using a preset objective function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, the first target model parameters, the second target model parameters, and the objective function value, calculate and determine the response characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model using a preset response evaluation function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, and the target operation data, use the first updated fuzzy control model and the second updated fuzzy control model to respectively determine the first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value and the second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value; Based on the first mechanical frequency change rate, the second mechanical frequency change rate, the first target model parameters, the second target model parameters, and the objective function value, calculate and determine the fluctuation characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model using a preset fluctuation evaluation function.

[0018] Further, when the model inference module is used to input the operation data into the target fuzzy control model corresponding to the disturbance state and perform inference calculation based on the operation data using the target fuzzy control model to obtain the target mechanical frequency output by the target fuzzy control model, the model inference module is used for: Input the operation data into the input layer set in the target fuzzy control model corresponding to the disturbance state to obtain the input vector output by the input layer; Input the input vector into the radial basis layer set in the target fuzzy control model, and use the radial basis layer to perform fuzzy processing on the input vector to obtain the radial basis output vector output by the radial basis layer; Input the radial basis output vector into the normalization layer set in the target fuzzy control model, and use the normalization layer to perform normalization processing on the radial basis output vector to obtain the normalized output vector output by the normalization layer; Input the normalized output vector into the output layer set in the target fuzzy control model to obtain the mechanical frequency change amount output by the output layer; Obtain the historical target mechanical frequency corresponding to the large rotating machine at the previous moment, and determine the sum of the historical target mechanical frequency and the mechanical frequency change amount as the target mechanical frequency output by the target fuzzy control model.

[0019] An embodiment of the present application further provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the control method for the large rotating machinery as described above are executed.

[0020] An embodiment of the present application further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps of the control method for the large rotating machinery as described above are executed.

[0021] The control method and device for large rotating machinery provided by the embodiment of the present application. The control method includes: during the operation of the large rotating machinery, obtaining the disturbance state and operation data corresponding to the large rotating machinery at the current moment; wherein, the disturbance state includes an external disturbance state and a non-external disturbance state; judging whether the cumulative operation data corresponding to the obtained operation data reaches a preset window amount, and based on the judgment result, determining the target fuzzy control model corresponding to the disturbance state; inputting the operation data into the target fuzzy control model corresponding to the disturbance state, and using the target fuzzy control model to perform inference calculation based on the operation data to obtain the target mechanical frequency output by the target fuzzy control model; controlling the motor arranged in the large rotating machinery to operate according to the target mechanical frequency.

[0022] Compared with the prior art methods such as PID control and adaptive control, by performing online parameter update and robustness evaluation on the fuzzy control model under different disturbance states, obtaining the target fuzzy control model under different disturbance states, and inputting the operation data of the large rotating machinery into the target fuzzy control model under the current disturbance state, controlling the large rotating machinery to operate according to the target mechanical frequency output by the target fuzzy control model, enabling the large rotating machinery to better cope with complex external environmental changes, improving the anti-interference ability under the external disturbance environment, improving the operation stability of the large rotating machinery, and further improving the safety of the large rotating machinery.

[0023] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, makes the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0025] Figure 1 It is a flowchart of a control method for a large rotating machine provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a control device for a large rotating machine provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present application. Usually, the components of the embodiments of the present application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, every other embodiment obtained by those of ordinary skill in the art without creative efforts belongs to the scope of protection of the present application.

[0027] Through research, it has been found that currently, control methods for large rotating machines usually adopt methods such as PID control and adaptive control. However, when these methods are used to control large rotating machines in a complex disturbance environment, problems such as poor anti-interference performance and difficulty in coping with rapidly changing disturbances will occur, reducing the stability of the operation of large rotating machines and further reducing the safety of large rotating machines.

[0028] Among them, in the method of controlling a large rotating machine based on PID control, the output signal of the large rotating machine is adjusted through three control parameters: proportional, integral, and differential, so as to achieve the control of the large rotating machine. However, due to the limited adaptability of PID control to a complex disturbance environment, it may not be able to maintain a stable control state under strong external disturbances. In addition, manual adjustment of parameters may also lead to problems with unstable performance of large rotating machines.

[0029] In addition, the method for adaptively controlling large rotating machinery adapts to changes in external disturbances by continuously adjusting the parameters of the controller, thereby improving the robustness of the system. However, when facing large rotating machinery systems with nonlinearity and uncertainty, the adaptive control method will reduce the stability and convergence of control.

[0030] Based on this, the embodiments of the present application provide a control method for large rotating machinery. By performing online parameter update and robustness evaluation on the fuzzy control model under different disturbance states, the target fuzzy control model under different disturbance states is obtained, and the operation data of the large rotating machinery is input into the target fuzzy control model under the current disturbance state to control the large rotating machinery to operate according to the target mechanical frequency output by the target fuzzy control model, enabling the large rotating machinery to better cope with complex changes in the external environment, improving the anti-interference ability under the external disturbance environment, enhancing the stability of the operation of the large rotating machinery, and further improving the safety of the large rotating machinery.

[0031] Please refer to Figure 1 , Figure 1 which is a flowchart of a control method for large rotating machinery provided by the embodiments of the present application. As Figure 1 shown in S100. During the operation of the large rotating machinery, obtain the disturbance state and operation data corresponding to the large rotating machinery at the current moment.

[0032] It should be noted that large rotating machinery refers to large equipment with key roles in industrial applications and including main rotating components, which are applied in multiple industries, including but not limited to the fields of energy, manufacturing, mining, and chemical engineering, etc. For example, large rotating machinery may include steam turbines, gas turbines, water turbines, wind turbines, rolling mills, centrifuges, etc.

[0033] Among them, the disturbance state includes an external disturbance state and a no-external-disturbance state.

[0034] Here, the external disturbance state indicates that the environment where the large rotating machinery is located has a situation that interferes with the operation of the large rotating machinery, and the no-external-disturbance state indicates that the environment where the large rotating machinery is located does not affect the operation of the large rotating machinery.

[0035] In this step, in specific implementation, first, use the environmental sensors set on the large rotating machinery to collect the environmental data corresponding to the environment where the large rotating machinery is located; then, based on this environmental data, determine the disturbance state corresponding to the large rotating machinery at the current moment; finally, obtain the operation data corresponding to the large rotating machinery at the current moment.

[0036] In the embodiments of the present application, the environmental data at least includes an air flow velocity value, a vibration frequency value, an environmental temperature value, etc.; the operation data at least includes a mechanical rotation wind speed value, an actual mechanical frequency value, an actual mechanical frequency deviation value, etc.

[0037] Wherein, the actual mechanical frequency deviation value is the difference between a preset desired frequency value and the actual mechanical frequency value.

[0038] S200. Determine whether the cumulative operation data corresponding to the obtained operation data reaches a preset window quantity, and based on the determination result, determine the target fuzzy control model corresponding to the disturbance state.

[0039] In this step, at the current moment, determine the cumulative data quantity of the operation data corresponding to the large rotating machinery obtained cumulatively, that is, determine the data quantity corresponding to the cumulative operation data corresponding to the obtained operation data, and determine whether the data quantity corresponding to the cumulative operation data corresponding to the obtained operation data reaches the preset window quantity, to obtain a determination result.

[0040] In the embodiments of the present application, the preset window quantity can be specifically calibrated according to the model architecture parameters of the fuzzy control model corresponding to each disturbance state and the actual control requirements.

[0041] Furthermore, based on the determination result obtained by determining whether the data quantity corresponding to the cumulative operation data corresponding to the obtained operation data reaches the preset window quantity, determine the target fuzzy control model corresponding to the disturbance state.

[0042] In the embodiments of the present application, each disturbance state corresponds to a fuzzy control model. This fuzzy control model can be updated online during the process of controlling the large rotating machinery. When corresponding conditions are met during the update process of this fuzzy control model, obtain the target fuzzy control model corresponding to each disturbance state, so as to use this target fuzzy control model to predict the target mechanical frequency of the large rotating machinery.

[0043] Here, the fuzzy control model can include a neural network model combining a fuzzy neural network and a radial basis function network. This fuzzy control model combines the computational intelligence technologies of a fuzzy logic system and a neural network model, aiming to utilize the ability of the fuzzy logic to process uncertain and imprecise information, as well as the advantages of the neural network in pattern recognition, learning, and adaptability. The fuzzy control model can handle complex, non-linear mapping problems and perform excellently when dealing with data with uncertainty or fuzziness.

[0044] Among them, the no external disturbance state corresponds to a first fuzzy control model and a first target fuzzy control model; the external disturbance state corresponds to a second fuzzy control model and a second target fuzzy control model.

[0045] In an implementation manner of the present application, during specific implementation, when the judgment result indicates that the cumulative operation data corresponding to the obtained operation data has not reached the preset window quantity, the step of determining the target fuzzy control model corresponding to the disturbance state in step S200 may include: S211. Obtain the fuzzy control model corresponding to the disturbance state updated online last time, and determine this fuzzy control model as the target fuzzy control model corresponding to the disturbance state.

[0046] In this step, when the judgment result indicates that the cumulative operation data corresponding to the obtained operation data has not reached the preset window quantity, it means that the cumulative quantity of the current operation data is not sufficient to update the fuzzy control model online. Determine the fuzzy control model corresponding to this disturbance state updated online last time as the target fuzzy control model corresponding to this disturbance state, that is, there is no need to update the model parameters of the fuzzy control model online.

[0047] In an implementation manner of the present application, during specific implementation, when the judgment result indicates that the cumulative operation data corresponding to the obtained operation data has reached the preset window quantity, the step of determining the target fuzzy control model corresponding to the disturbance state in step S200 may include: S221. Obtain the fuzzy control model corresponding to the disturbance state updated online last time, and use the cumulative operation data to update the model parameters in the fuzzy control model to obtain the updated fuzzy control model corresponding to the disturbance state.

[0048] In this step, according to the characteristics that large rotating machinery is vulnerable to external interference, a target function for controlling the robustness of the mechanical frequency in large rotating machinery is preset, and the adaptive gradient descent method is used to update the model parameters in the fuzzy control model online.

[0049] In the embodiments of the present application, the adaptive gradient descent algorithm is used to dynamically update the model parameters in the fuzzy control model, which improves the learning ability and calculation efficiency of the fuzzy control model. By optimizing the learning rate of the fuzzy control model, the fuzzy control model can more quickly adapt to the changes of large rotating machinery, and improves the operation stability of controlling large rotating machinery.

[0050] In an implementation manner of the present application, during specific implementation, when the disturbance state is a state without external disturbance, the step of using the cumulative operation data to update the model parameters in the fuzzy control model to obtain the updated fuzzy control model corresponding to the disturbance state in step S221 may include: S22111. Input the first cumulative operation data in the cumulative operation data into the first fuzzy control model corresponding to the no-external-disturbance state to obtain the first mechanical frequency value output by the first fuzzy control model.

[0051] In the embodiments of the present application, the fuzzy control model includes an input layer, a radial basis layer, a normalization layer, and an output layer.

[0052] In this step, screen out the first cumulative operation data corresponding to the first moment in chronological order from the cumulative operation data, and input the first cumulative operation data into the first fuzzy control model corresponding to the no-external-disturbance state; then, through the calculations of the input layer, radial basis layer, normalization layer, and output layer of the first fuzzy control model, obtain the first mechanical frequency value output by the first fuzzy control model based on the first cumulative operation data.

[0053] S22112. Based on the first mechanical frequency value and the first preset mechanical frequency value, use the first objective function preset corresponding to the no-external-disturbance state to calculate and determine the first target value corresponding to the first cumulative operation data in the no-external-disturbance state.

[0054] In the embodiments of the present application, the expression of the first objective function preset corresponding to the no-external-disturbance state is as follows.

[0055] 。

[0056] Wherein, represents the first target value corresponding to the no-external-disturbance state; represents the first mechanical frequency value; represents the first preset mechanical frequency value.

[0057] S22113. Determine whether the first target value is less than the preset target value.

[0058] In the embodiments of the present application, the preset target value can generally be specifically calibrated according to the convergence requirements of the objective function for online updating model parameters, generally set to 0.01, or can also be set to other values, which are not limited herein in the present application.

[0059] S22114. If not, then based on the first target value, perform an update calculation on the model parameters in the first fuzzy control model to obtain the first model parameters corresponding to the first update period, and replace the model parameters in the first fuzzy control model with the first model parameters to obtain the first fuzzy control model corresponding to the first update period.

[0060] In this step, when the first target value is greater than or equal to the preset target value, the model parameters in the first fuzzy control model are updated and calculated using a preset update formula to obtain the first model parameters corresponding to the first update period representing this online update; then, the model parameters in the first fuzzy control model are replaced with the first model parameters to obtain the first fuzzy control model corresponding to the first update period.

[0061] Among them, the model parameters in the fuzzy control model at least include the central value and width value corresponding to the input neuron and the radial basis neuron, and the output weight corresponding to the normalization neuron.

[0062] In the embodiment of the present application, the expression of the update formula for updating and calculating the model parameters in the first fuzzy control model is as follows.

[0063] 。

[0064] 。

[0065] 。

[0066] Among them, represents the central value corresponding to the input neuron and the radial basis neuron in the first model parameters; represents the width value corresponding to the input neuron and the radial basis neuron in the first model parameters; represents the output weight corresponding to the normalization neuron in the first model parameters; represents the central value corresponding to the input neuron and the radial basis neuron in the model parameters; represents the width value corresponding to the input neuron and the radial basis neuron in the model parameters; represents the output weight corresponding to the normalization neuron in the model parameters; represents the first target value corresponding to the state without external disturbance; represents For the partial derivative of; represents For the partial derivative of; represents For the partial derivative of.

[0067] S22115. Repeatedly input the cumulative operation data into the first fuzzy control model corresponding to each update period, calculate the first target value corresponding to each update period, and when the first target value is greater than or equal to the preset target value, update the model parameters in the first fuzzy control model to obtain the first model parameters corresponding to each update period until the first target value is less than the preset target value.

[0068] In this step, after obtaining the first fuzzy control model corresponding to the first update period, input the second cumulative operation data corresponding to the second moment in chronological order in the cumulative operation data into the first fuzzy control model corresponding to the first update period, calculate the first target value corresponding to the next update period, and determine whether the first target value is less than the preset target value. When the first target value is greater than or equal to the preset target value, update the model parameters in the first fuzzy control model.

[0069] Further, when the first target value is greater than or equal to the preset target value, repeatedly execute the steps of updating the model parameters in the first fuzzy control model and inputting the cumulative operation data into the first fuzzy control model corresponding to each update period until the first target value is less than the preset target value.

[0070] S22116. When the first target value is less than the preset target value, determine the first target model parameters corresponding to the current update period, and replace the model parameters in the first fuzzy control model with the first target model parameters to obtain the first updated fuzzy control model corresponding to the state without external disturbance.

[0071] In this step, when the first target value corresponding to the target update period is less than the preset target value, determine the first target model parameters corresponding to the updated target update period (the current update period); then, replace the model parameters in the first fuzzy control model with the first target model parameters to obtain the first updated fuzzy control model corresponding to the state without external disturbance.

[0072] In an embodiment of the present application, in specific implementation, when the disturbance state is an external disturbance state, the step of using the cumulative operation data to online update the model parameters in the fuzzy control model to obtain the updated fuzzy control model corresponding to the disturbance state in step S221 may include: S22121. Input the first cumulative operation data in the cumulative operation data into the second fuzzy control model corresponding to the external disturbance state to obtain the second mechanical frequency value output by the second fuzzy control model.

[0073] In this step, the first cumulative operation data corresponding to the first moment in the chronological order is screened out from the cumulative operation data, and the first cumulative operation data is input into the second fuzzy control model corresponding to the external disturbance state; then, through the calculations of the input layer, radial basis layer, normalization layer and output layer of the second fuzzy control model, the second mechanical frequency value output by the second fuzzy control model based on the first cumulative operation data is obtained.

[0074] S22122. Based on the second mechanical frequency value, the second preset mechanical frequency value, the first target model parameters and the model parameters in the second fuzzy control model, use the second target function preset corresponding to the external disturbance state to calculate and determine the second target value corresponding to the first cumulative operation data under the external disturbance state.

[0075] In the embodiment of the present application, the expression of the second target function preset corresponding to the external disturbance state is as follows.

[0076] .

[0077] Wherein, represents the second target value corresponding to the external disturbance state; represents the second mechanical frequency value; represents the second preset mechanical frequency value; represents the first target model parameters in the first fuzzy control model; represents the model parameters in the second fuzzy control model; represents the factorial function.

[0078] S22123. Judge whether the second target value is less than the preset target value.

[0079] S22124. If not, based on the second target value, perform an update calculation on the model parameters in the second fuzzy control model to obtain the second model parameters corresponding to the first update period, and replace the model parameters in the second fuzzy control model with the second model parameters to obtain the second fuzzy control model corresponding to the first update period.

[0080] S22125. Repeatedly input the cumulative operation data into the second fuzzy control model corresponding to each update period, calculate the second target value corresponding to each update period, and when the second target value is greater than or equal to the preset target value, update the model parameters in this second fuzzy control model to obtain the second model parameters corresponding to each update period until the second target value is less than the preset target value.

[0081] S22126. When the second target value is less than the preset target value, determine the second target model parameters corresponding to the current update period, and replace the model parameters in the second fuzzy control model with the second target model parameters to obtain the second updated fuzzy control model corresponding to the external disturbance state.

[0082] Among them, the descriptions of S22123 to S22126 can refer to the descriptions of S22113 to S22116, and the same technical effects can be achieved, so details are not described herein.

[0083] S222. Use the target operation data corresponding to when the updated fuzzy control model is obtained to perform a robustness evaluation on the updated fuzzy control model to obtain a robustness evaluation result, and based on the robustness evaluation result, adjust the updated fuzzy control model corresponding to each disturbance state to obtain the target fuzzy control model corresponding to the disturbance state.

[0084] In the embodiments of the present application, a robustness evaluation strategy is used to analyze and evaluate the dynamic characteristics of the updated fuzzy control model, thereby enhancing the anti-interference ability of large rotating machinery, enabling better response to complex external environmental changes when controlling large rotating machinery, and improving the operation stability of large rotating machinery.

[0085] In an implementation manner of the present application, in specific implementation, the step of using the target operation data corresponding to when the updated fuzzy control model is obtained to perform a robustness evaluation on the updated fuzzy control model to obtain a robustness evaluation result in step S222 may include: S2221. Input the target operation data corresponding to when the updated fuzzy control model is obtained into the first updated fuzzy control model corresponding to the state without external disturbance and the second updated fuzzy control model corresponding to the external disturbance state respectively, and obtain a first evaluated mechanical frequency value output by the first updated fuzzy control model and a second evaluated mechanical frequency value output by the second updated fuzzy control model respectively.

[0086] In this step, in specific implementation, first, when the updated fuzzy control model is obtained, determine the first target operation data input to the first updated fuzzy control model corresponding to the state without external disturbance and making the target value less than the preset target value, and determine the second target operation data input to the second updated fuzzy control model corresponding to the external disturbance state and making the target value less than the preset target value; then, input the first target operation data and the second target operation data into the first updated fuzzy control model and the second updated fuzzy control model respectively; finally, obtain a first evaluated mechanical frequency value output by the first updated fuzzy control model and a second evaluated mechanical frequency value output by the second updated fuzzy control model.

[0087] S2222. Calculate and determine the objective function value by using a preset objective function based on the first target model parameter in the first updated fuzzy control model, the second target model parameter in the second updated fuzzy control model, the first evaluated mechanical frequency value, and the second evaluated mechanical frequency value.

[0088] In this step, in specific implementation, first, calculate and determine the first objective function value corresponding to the first updated fuzzy control model by using the first objective function based on the first evaluated mechanical frequency value and the first preset mechanical frequency value; then, calculate and determine the second objective function value corresponding to the second updated fuzzy control model by using the second objective function based on the second evaluated mechanical frequency value, the second preset mechanical frequency value, the first target model parameter, and the second model parameter; finally, integrate the first objective function value and the second objective function value into the objective function value.

[0089] In the embodiment of the present application, the expression of the objective function value is as follows.

[0090] ∈( , )。

[0091] Wherein, represents the objective function value; represents the first objective function value; represents the second objective function value.

[0092] S2223. Calculate and determine the response characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model by using a preset response evaluation function based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, the first target model parameter, the second target model parameter, and the objective function value.

[0093] In the embodiment of the present application, the expression of the response evaluation function is as follows.

[0094] 。

[0095] Wherein, represents the response characteristic result in the robustness evaluation result; represents the first evaluated mechanical frequency value; represents the second evaluated mechanical frequency value; represents the objective function value; represents the parameter matrix composed of the first target model parameter and the second target model parameter; e is the natural constant.

[0096] S2224. Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, and the target operation data, use the first updated fuzzy control model and the second updated fuzzy control model to respectively determine the first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value and the second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value.

[0097] In this step, based on the first evaluated mechanical frequency value and the target operation data, use the first updated fuzzy control model to determine the first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value; then, based on the second evaluated mechanical frequency value and the target operation data, use the second updated fuzzy control model to determine the second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value.

[0098] S2225. Based on the first mechanical frequency change rate, the second mechanical frequency change rate, the first target model parameter, the second target model parameter, and the target function value, use a preset fluctuation evaluation function to calculate and determine the fluctuation characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model.

[0099] In the application embodiment, the expression of the fluctuation evaluation function is as follows.

[0100] .

[0101] Wherein, represents the fluctuation characteristic result in the robustness evaluation result; represents the first mechanical frequency change rate; represents the second mechanical frequency change rate; represents the target function value; represents the parameter matrix composed of the first target model parameter and the second target model parameter.

[0102] S300. Input the operation data into the target fuzzy control model corresponding to the perturbation state, and use the target fuzzy control model to perform inference calculation based on the operation data to obtain the target mechanical frequency output by the target fuzzy control model.

[0103] It should be noted that the first target fuzzy control model corresponding to the state without external perturbation and the second target fuzzy control model corresponding to the state with external perturbation both include the same fuzzy control neural network architecture. There are differences in the training data used in the training processes of the first target fuzzy control model and the second target fuzzy control model, which represent the presence or absence of external interference, resulting in differences in the model parameters of the first target fuzzy control model and the second target fuzzy control model. The first target fuzzy control model and the second target fuzzy control model perform corresponding inference calculations for the state without external perturbation and the state with external perturbation respectively.

[0104] In the embodiment of the present application, based on the target fuzzy control model corresponding to each perturbation state, the target mechanical frequency is calculated through the variables in the dynamic inference model, ensuring the safety and stability of the operation of large rotating machinery, enabling large rotating machinery to more flexibly respond to an unstable working environment, and achieving efficient control of large rotating machinery.

[0105] In an implementation manner of the present application, in specific implementation, step S300 may include: S301. Input the operation data into the input layer set in the target fuzzy control model corresponding to the perturbation state, and obtain the input vector output by the input layer.

[0106] In this step, the mechanical rotation wind speed value and the actual mechanical frequency deviation value included in the operation data are input into the input layer set in the obtained target fuzzy control model corresponding to the perturbation state, and the input vector output by the input layer is obtained.

[0107] In the embodiment of the present application, the expression of the input vector output by the input layer in the target fuzzy control model corresponding to the no-external-perturbation state is as follows.

[0108]

[0109] Among them, represents the input vector output by the input layer in the target fuzzy control model corresponding to the no-external-perturbation state at the current moment; represents the mechanical rotation wind speed value corresponding to the no-external-perturbation state at the current moment; represents the actual mechanical frequency deviation value corresponding to the no-external-perturbation state at the current moment.

[0110] In the embodiment of the present application, the expression of the input vector output by the input layer in the target fuzzy control model corresponding to the external-perturbation state is as follows.

[0111] .

[0112] Among them, represents the input vector output by the input layer in the target fuzzy control model corresponding to the external-perturbation state at the current moment; represents the mechanical rotation wind speed value corresponding to the external-perturbation state at the current moment; represents the actual mechanical frequency deviation value corresponding to the external-perturbation state at the current moment.

[0113] S302. Input the input vector into the radial basis layer set in the target fuzzy control model, and use the radial basis layer to perform fuzzy processing on the input vector to obtain the radial basis output vector output by the radial basis layer.

[0114] Among them, the radial basis layer includes a plurality of radial basis neurons. For example, the number of radial basis neurons can be any positive integer selected from [5, 10].

[0115] In the embodiment of the present application, the expression of the radial basis output vector output by the radial basis layer in the target fuzzy control model corresponding to the state without external disturbance is as follows.

[0116] 。

[0117] Among them, represents the radial basis output value output by each radial basis neuron in the radial basis layer in the target fuzzy control model corresponding to the state without external disturbance, that is, the radial basis output vector output by the radial basis layer in the target fuzzy control model corresponding to the state without external disturbance can be determined; represents the operation data corresponding to the state without external disturbance at the current moment; represents the th input neuron and the th radial basis neuron in the target fuzzy control model corresponding to the state without external disturbance; represents the th input neuron and the th radial basis neuron in the target fuzzy control model corresponding to the state without external disturbance; k represents the number of input neurons in the input layer of the target fuzzy control model corresponding to the state without external disturbance.

[0118] In the embodiment of the present application, the expression of the radial basis output vector output by the radial basis layer in the target fuzzy control model corresponding to the state with external disturbance is as follows.

[0119]

[0120] Among them, represents the radial basis output value output by each radial basis neuron in the radial basis layer in the target fuzzy control model corresponding to the state with external disturbance, that is, the radial basis output vector output by the radial basis layer in the target fuzzy control model corresponding to the state with external disturbance can be determined; represents the operation data corresponding to the state with external disturbance at the current moment; represents the th input neuron and the th radial basis neuron in the target fuzzy control model corresponding to the state with external disturbance; represents the th input neuron and the The corresponding width value between radial basis neurons.

[0121] S303. Input the radial basis output vector into the normalization layer set in the target fuzzy control model, and use the normalization layer to perform normalization processing on the radial basis output vector to obtain the normalized output vector output by the normalization layer.

[0122] Among them, the normalization layer includes multiple normalization neurons.

[0123] In the embodiment of the present application, the expression of the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state without external disturbance is as follows.

[0124] .

[0125] Among them, represents the normalized output value output by each normalization neuron in the normalization layer in the target fuzzy control model corresponding to the state without external disturbance, and the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state without external disturbance can be determined; represents the operation data corresponding to the state without external disturbance at the current moment; represents the th input neuron and the th corresponding center value between the radial basis neurons in the target fuzzy control model corresponding to the state without external disturbance; represents the th input neuron and the th corresponding width value between the radial basis neurons in the target fuzzy control model corresponding to the state without external disturbance, l represents the number of radial basis neurons in the radial basis layer of the target fuzzy control model corresponding to the state without external disturbance.

[0126] In the embodiment of the present application, the expression of the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state with external disturbance is as follows.

[0127] .

[0128] Among them, represents the normalized output value output by each normalization neuron in the normalization layer in the target fuzzy control model corresponding to the state with external disturbance, and the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state with external disturbance can be determined; represents the operation data corresponding to the state with external disturbance at the current moment; represents the The central value corresponding to the th input neuron and the th radial basis neuron; The width value corresponding to the th input neuron and the

[0129] S304. Input the normalized output vector into the output layer set in the target fuzzy control model to obtain the mechanical frequency change amount output by the output layer.

[0130] In the embodiment of the present application, the expression of the mechanical frequency change amount output by the output layer in the target fuzzy control model corresponding to the state without external disturbance is as follows.

[0131] .

[0132] Wherein, represents the mechanical frequency change amount output by the output layer in the target fuzzy control model corresponding to the state without external disturbance; represents the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state without external disturbance; represents the output weight corresponding to the normalization layer in the target fuzzy control model corresponding to the state without external disturbance.

[0133] In the embodiment of the present application, the expression of the mechanical frequency change amount output by the output layer in the target fuzzy control model corresponding to the state with external disturbance is as follows.

[0134] .

[0135] Wherein, represents the mechanical frequency change amount output by the output layer in the target fuzzy control model corresponding to the state with external disturbance; represents the normalized output vector output by the normalization layer in the target fuzzy control model corresponding to the state with external disturbance; represents the output weight corresponding to the normalization layer in the target fuzzy control model corresponding to the state with external disturbance.

[0136] S305. Obtain the historical target mechanical frequency corresponding to the large rotating machinery at the previous moment, and determine the sum of the historical target mechanical frequency and the mechanical frequency change amount as the target mechanical frequency output by the target fuzzy control model.

[0137] In the embodiment of the present application, the expression of the target mechanical frequency output by the target fuzzy control model corresponding to the state without external disturbance is as follows.

[0138] 。

[0139] Among them, represents the target mechanical frequency output by the target fuzzy control model corresponding to the state without external disturbance; represents the historical target mechanical frequency corresponding to the state without external disturbance; represents the change in mechanical frequency output by the output layer in the target fuzzy control model corresponding to the state without external disturbance.

[0140] In the embodiment of the present application, the expression of the target mechanical frequency output by the target fuzzy control model corresponding to the external disturbance state is as follows.

[0141] 。

[0142] Among them, represents the target mechanical frequency output by the target fuzzy control model corresponding to the external disturbance state; represents the historical target mechanical frequency corresponding to the external disturbance state; represents the change in mechanical frequency output by the output layer in the target fuzzy control model corresponding to the external disturbance state.

[0143] S400. Control the motor disposed in the large rotating machine to operate at the target mechanical frequency.

[0144] In this step, when obtaining the target mechanical frequency corresponding to the disturbance state at the current moment, input the target mechanical frequency into the frequency converter in the large rotating machine, and control the frequency converter to adjust the speed of the motor in the large rotating machine according to the target mechanical frequency, so that the mechanical frequency of the large rotating machine reaches the target mechanical frequency, and further enables the large rotating machine to operate at the target mechanical frequency.

[0145] The control method of the large rotating machine provided by the embodiment of the present application obtains the target fuzzy control model under different disturbance states by performing online parameter update and robustness evaluation on the fuzzy control model under different disturbance states, and inputs the operation data of the large rotating machine into the target fuzzy control model under the current disturbance state, and controls the large rotating machine to operate at the target mechanical frequency output by the target fuzzy control model, so that the large rotating machine can better cope with complex external environmental changes, improve the anti-interference ability under the external disturbance environment, improve the operation stability of the large rotating machine, and further improve the safety of the large rotating machine.

[0146] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of a control device for a large rotating machine provided by an embodiment of the present application. As shown in Figure 2 shown in, the control device 200 includes: A data acquisition module 210, configured to acquire the disturbance state and operation data corresponding to the large rotating machinery at the current moment during the operation of the large rotating machinery; wherein, the disturbance state includes an external disturbance state and a non-external disturbance state; A parameter update module 220, configured to determine whether the cumulative operation data corresponding to the acquired operation data reaches a preset window quantity, and based on the determination result, determine the target fuzzy control model corresponding to the disturbance state; A model inference module 230, configured to input the operation data into the target fuzzy control model corresponding to the disturbance state, and use the target fuzzy control model to perform inference calculation based on the operation data to obtain the target mechanical frequency output by the target fuzzy control model; A mechanical control module 240, configured to control the motor disposed in the large rotating machinery to operate at the target mechanical frequency.

[0147] Further, when the determination result is that the cumulative operation data corresponding to the acquired operation data does not reach the preset window quantity, when the parameter update module 220 is used to determine the target fuzzy control model corresponding to the disturbance state based on the determination result, the parameter update module 220 is configured to: Acquire the fuzzy control model corresponding to the disturbance state updated last time online, and determine the fuzzy control model as the target fuzzy control model corresponding to the disturbance state.

[0148] Further, when the determination result is that the cumulative operation data corresponding to the acquired operation data reaches the preset window quantity, when the parameter update module 220 is used to determine the target fuzzy control model corresponding to the disturbance state based on the determination result, the parameter update module 220 is configured to: Acquire the fuzzy control model corresponding to the disturbance state updated last time online, and use the cumulative operation data to update the model parameters in the fuzzy control model online to obtain the updated fuzzy control model corresponding to the disturbance state; Use the target operation data corresponding to the obtained updated fuzzy control model to perform a robustness evaluation on the updated fuzzy control model to obtain a robustness evaluation result, and based on the robustness evaluation result, adjust the updated fuzzy control model corresponding to each disturbance state to obtain the target fuzzy control model corresponding to the disturbance state.

[0149] Further, when the disturbance state is a non-external disturbance state, when the parameter update module 220 is used to update the model parameters in the fuzzy control model online by using the acquired cumulative operation data to obtain the updated fuzzy control model corresponding to the disturbance state, the parameter update module 220 is configured to: Input the first cumulative operation data in the cumulative operation data into the first fuzzy control model corresponding to the no-external-disturbance state to obtain a first mechanical frequency value output by the first fuzzy control model; Based on the first mechanical frequency value and the first preset mechanical frequency value, use the first objective function preset corresponding to the no-external-disturbance state to calculate and determine a first objective value corresponding to the first cumulative operation data in the no-external-disturbance state; Judge whether the first objective value is less than a preset objective value; If not, then based on the first objective value, perform an update calculation on the model parameters in the first fuzzy control model to obtain first model parameters corresponding to a first update period, and replace the model parameters in the first fuzzy control model with the first model parameters to obtain a first fuzzy control model corresponding to the first update period; Repeat inputting the cumulative operation data into the first fuzzy control model corresponding to each update period, calculate the first objective value corresponding to each update period, and when the first objective value is greater than or equal to the preset objective value, update the model parameters in the first fuzzy control model to obtain first model parameters corresponding to each update period until the first objective value is less than the preset objective value; When the first objective value is less than the preset objective value, determine first target model parameters corresponding to the current update period, and replace the model parameters in the first fuzzy control model with the first target model parameters to obtain a first updated fuzzy control model corresponding to the no-external-disturbance state.

[0150] Further, when the disturbance state is an external disturbance state, when the parameter update module 220 is used to online update the model parameters in the fuzzy control model by using the operation data to obtain an updated fuzzy control model corresponding to the disturbance state, the parameter update module 220 is used for: Input the first cumulative operation data in the cumulative operation data into the second fuzzy control model corresponding to the external disturbance state to obtain a second mechanical frequency value output by the second fuzzy control model; Based on the second mechanical frequency value, the second preset mechanical frequency value, the first target model parameters and the model parameters in the second fuzzy control model, use the second objective function preset corresponding to the external disturbance state to calculate and determine a second objective value corresponding to the first cumulative operation data in the external disturbance state; Judge whether the second objective value is less than the preset objective value; Otherwise, based on the second target value, update and calculate the model parameters in the second fuzzy control model to obtain the second model parameters corresponding to the first update period, and replace the model parameters in the second fuzzy control model with the second model parameters to obtain the second fuzzy control model corresponding to the first update period; Repeat inputting the cumulative operation data into the second fuzzy control model corresponding to each update period, calculate the second target value corresponding to each update period, and when the second target value is greater than or equal to the preset target value, update the model parameters in the second fuzzy control model to obtain the second model parameters corresponding to each update period until the second target value is less than the preset target value; When the second target value is less than the preset target value, determine the second target model parameters corresponding to the current update period, and replace the model parameters in the second fuzzy control model with the second target model parameters to obtain the second updated fuzzy control model corresponding to the external disturbance state.

[0151] Further, when the parameter update module 220 is used to evaluate the robustness of the updated fuzzy control model by using the target operation data corresponding to the obtained updated fuzzy control model to obtain the robustness evaluation result, the parameter update module 220 is used for: Input the target operation data corresponding to the obtained updated fuzzy control model into the first updated fuzzy control model corresponding to the no-external-disturbance state and the second updated fuzzy control model corresponding to the external disturbance state respectively, and obtain the first evaluated mechanical frequency value output by the first updated fuzzy control model and the second evaluated mechanical frequency value output by the second updated fuzzy control model respectively; Based on the first target model parameters in the first updated fuzzy control model, the second target model parameters in the second updated fuzzy control model, the first evaluated mechanical frequency value, and the second evaluated mechanical frequency value, calculate and determine the objective function value by using a preset objective function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, the first target model parameters, the second target model parameters, and the objective function value, calculate and determine the response characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model by using a preset response evaluation function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value, and the target operation data, use the first updated fuzzy control model and the second updated fuzzy control model to respectively determine the first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value and the second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value; Based on the first mechanical frequency change rate, the second mechanical frequency change rate, the first target model parameter, the second target model parameter, and the objective function value, use a preset fluctuation evaluation function to calculate and determine the fluctuation characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model.

[0152] Further, when the model inference module 230 is used to input the operation data into the target fuzzy control model corresponding to the perturbation state and perform inference calculation based on the operation data by using the target fuzzy control model to obtain the target mechanical frequency output by the target fuzzy control model, the model inference module 230 is used for: Input the operation data into the input layer set in the target fuzzy control model corresponding to the perturbation state to obtain an input vector output by the input layer; Input the input vector into the radial basis layer set in the target fuzzy control model, and use the radial basis layer to perform fuzzy processing on the input vector to obtain a radial basis output vector output by the radial basis layer; Input the radial basis output vector into the normalization layer set in the target fuzzy control model, and use the normalization layer to perform normalization processing on the radial basis output vector to obtain a normalized output vector output by the normalization layer; Input the normalized output vector into the output layer set in the target fuzzy control model to obtain a mechanical frequency change amount output by the output layer; Obtain the historical target mechanical frequency corresponding to the large rotating machine at the previous moment, and determine the sum of the historical target mechanical frequency and the mechanical frequency change amount as the target mechanical frequency output by the target fuzzy control model.

[0153] The control device for a large rotating machine provided by the embodiments of the present application obtains target fuzzy control models in different perturbation states by performing online parameter update and robustness evaluation on the fuzzy control models in different perturbation states, and inputs the operation data of the large rotating machine into the target fuzzy control model in the current perturbation state to control the large rotating machine to operate according to the target mechanical frequency output by the target fuzzy control model, so that the large rotating machine can better cope with complex external environmental changes, improve the anti-interference ability in the external perturbation environment, improve the operation stability of the large rotating machine, and further improve the safety of the large rotating machine.

[0154] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of an electronic device provided by the embodiments of the present application. As Figure 3 shown in, the electronic device 300 includes a processor 310, a memory 320, and a bus 330.

[0155] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 runs, the processor 310 communicates with the memory 320 via the bus 330. When the machine-readable instructions are executed by the processor 310, the steps of the control method for large rotating machinery in the method embodiment as described above can be executed. For the specific implementation manner, reference can be made to the method embodiment and will not be elaborated herein. Figure 1 shown in the method embodiment, and for the specific implementation manner, reference can be made to the method embodiment and will not be elaborated herein.

[0156] The embodiment of the present application also provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps of the control method for large rotating machinery in the method embodiment as described above can be executed. For the specific implementation manner, reference can be made to the method embodiment and will not be elaborated herein. Figure 1 shown in the method embodiment, and for the specific implementation manner, reference can be made to the method embodiment and will not be elaborated herein.

[0157] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0158] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings, direct couplings, or communication connections shown or discussed with each other can be through some communication interfaces. The indirect couplings or communication connections between devices or units can be in electrical, mechanical, or other forms.

[0159] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0161] When the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0162] Finally, it should be noted that the above-mentioned embodiments are only specific implementation manners of the present application, used to illustrate the technical solutions of the present application, rather than limiting it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions described in the foregoing embodiments or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A control method for a large rotating machine, characterized in that: The control method comprises: During the operation of the large-scale rotating machinery, the disturbance state and operation data corresponding to the large-scale rotating machinery at the current moment are obtained; wherein the disturbance state includes an external disturbance state and a state without external disturbance; Determine whether the accumulated operation data corresponding to the acquired operation data reaches a preset window amount, and determine the target fuzzy control model corresponding to the disturbance state based on the determination result; Inputting the operating data into a target fuzzy control model corresponding to the disturbance state, and using the target fuzzy control model to perform inference calculation based on the operating data to obtain a target mechanical frequency output by the target fuzzy control model; An electric motor installed in the large rotating machine is controlled to operate according to the target machine frequency.

2. The method according to claim 1, characterized in that When the judgment result is that the accumulated operation data corresponding to the acquired operation data does not reach the preset window amount, the target fuzzy control model corresponding to the disturbance state is determined based on the judgment result, including: A fuzzy control model corresponding to the disturbance state updated online last time is obtained, and the fuzzy control model is determined as a target fuzzy control model corresponding to the disturbance state.

3. The method according to claim 1, characterized in that: When the judgment result is that the accumulated operation data corresponding to the acquired operation data reaches a preset window amount, determining the target fuzzy control model corresponding to the disturbance state based on the judgment result includes: Acquire the fuzzy control model corresponding to the disturbance state updated online last time, and use the accumulated operation data to update the model parameters in the fuzzy control model online to obtain the updated fuzzy control model corresponding to the disturbance state; The updated fuzzy control model is evaluated for robustness using the target operating data corresponding to the updated fuzzy control model to obtain a robustness evaluation result, and based on the robustness evaluation result, the updated fuzzy control model corresponding to each disturbance state is adjusted to obtain a target fuzzy control model corresponding to the disturbance state.

4. The method according to claim 3, characterized in that When the disturbance state is a state without external disturbance, the online updating of the model parameters in the fuzzy control model using the acquired accumulated operation data to obtain an updated fuzzy control model corresponding to the disturbance state includes: Inputting first accumulated operation data in the accumulated operation data into a first fuzzy control model corresponding to the state without external disturbance to obtain a first mechanical frequency value output by the first fuzzy control model; Based on the first mechanical frequency value and the first preset mechanical frequency value, a first target value corresponding to the state without external disturbance is calculated and determined by using a preset first target function corresponding to the state without external disturbance; Determining whether the first target value is less than a preset target value; If not, then based on the first target value, the model parameters in the first fuzzy control model are updated and calculated to obtain the first model parameters corresponding to the first update period, and the model parameters in the first fuzzy control model are replaced with the first model parameters to obtain the first fuzzy control model corresponding to the first update period; Repeatingly inputting the accumulated operation data into the first fuzzy control model corresponding to each update period, calculating the first target value corresponding to each update period, and updating the model parameters in the first fuzzy control model when the first target value is greater than or equal to a preset target value, to obtain the first model parameters corresponding to each update period, until the first target value is less than the preset target value; When the first target value is less than the preset target value, the first target model parameters corresponding to the current update cycle are determined, and the model parameters in the first fuzzy control model are replaced with the first target model parameters to obtain the first updated fuzzy control model corresponding to the state without external disturbance.

5. The method according to claim 4, characterized in that When the disturbance state is an external disturbance state, the online updating of the model parameters in the fuzzy control model using the operation data to obtain an updated fuzzy control model corresponding to the disturbance state includes: inputting first accumulated operation data in the accumulated operation data into a second fuzzy control model corresponding to the external disturbance state to obtain a second mechanical frequency value output by the second fuzzy control model; Based on the second mechanical frequency value, the second preset mechanical frequency value, the first target model parameter and the model parameter in the second fuzzy control model, a second target value corresponding to the first accumulated operating data under the external disturbance state is calculated and determined by using a second target function preset corresponding to the external disturbance state; Determining whether the second target value is less than a preset target value; If not, then based on the second target value, the model parameters in the second fuzzy control model are updated and calculated to obtain the second model parameters corresponding to the first update period, and the model parameters in the second fuzzy control model are replaced with the second model parameters to obtain the second fuzzy control model corresponding to the first update period; Repeatingly inputting the accumulated operation data into the second fuzzy control model corresponding to each update period, calculating the second target value corresponding to each update period, and updating the model parameters in the second fuzzy control model when the second target value is greater than or equal to the preset target value, to obtain the second model parameters corresponding to each update period, until the second target value is less than the preset target value; When the second target value is less than the preset target value, the second target model parameters corresponding to the current update cycle are determined, and the model parameters in the second fuzzy control model are replaced with the second target model parameters to obtain a second updated fuzzy control model corresponding to the external disturbance state.

6. The method according to claim 3, characterized in that The step of performing robustness evaluation on the updated fuzzy control model by using the target operation data corresponding to the updated fuzzy control model to obtain a robustness evaluation result includes: Inputting the target operation data corresponding to the updated fuzzy control model into the first updated fuzzy control model corresponding to the state without external disturbance and the second updated fuzzy control model corresponding to the state with external disturbance, respectively, to obtain the first estimated mechanical frequency value output by the first updated fuzzy control model and the second estimated mechanical frequency value output by the second updated fuzzy control model; Determine an objective function value by using a preset objective function calculation based on a first objective model parameter in the first updated fuzzy control model, a second objective model parameter in the second updated fuzzy control model, the first evaluated mechanical frequency value, and the second evaluated mechanical frequency value; Based on the first evaluation mechanical frequency value, the second evaluation mechanical frequency value, the first target model parameter, the second target model parameter and the objective function value, a response characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model is calculated and determined using a preset response evaluation function; Based on the first evaluated mechanical frequency value, the second evaluated mechanical frequency value and the target operation data, using the first updated fuzzy control model and the second updated fuzzy control model to respectively determine a first mechanical frequency change rate corresponding to the first evaluated mechanical frequency value and a second mechanical frequency change rate corresponding to the second evaluated mechanical frequency value; Based on the first mechanical frequency change rate, the second mechanical frequency change rate, the first target model parameter, the second target model parameter and the objective function value, a preset fluctuation evaluation function is used to calculate and determine the fluctuation characteristic result in the robustness evaluation result corresponding to the updated fuzzy control model.

7. The method according to claim 1, characterized in that The step of inputting the operating data into a target fuzzy control model corresponding to the disturbance state, and using the target fuzzy control model to perform inference calculation based on the operating data to obtain a target mechanical frequency output by the target fuzzy control model includes: Inputting the operating data into an input layer set by a target fuzzy control model corresponding to the disturbance state to obtain an input vector output by the input layer; Inputting the input vector into the radial basis set by the target fuzzy control model, using the radial basis to perform fuzzy processing on the input vector, and obtaining a radial basis output vector output by the radial basis; Inputting the radial basis output vector into a normalization layer set in the target fuzzy control model, and using the normalization layer to normalize the radial basis output vector to obtain a normalized output vector output by the normalization layer; Inputting the normalized output vector into the output layer set by the target fuzzy control model to obtain the mechanical frequency change output by the output layer; The historical target mechanical frequency corresponding to the large rotating machine at the previous moment is obtained, and the sum of the historical target mechanical frequency and the mechanical frequency change is determined as the target mechanical frequency output by the target fuzzy control model.

8. A control device for a large rotating machine, characterized in that: The control device comprises: A data acquisition module is used to acquire the disturbance state and operation data corresponding to the large-scale rotating machinery at the current moment during the operation of the large-scale rotating machinery; wherein the disturbance state includes an external disturbance state and a state without external disturbance; A parameter updating module, used for judging whether the accumulated operation data corresponding to the acquired operation data reaches a preset window amount, and determining a target fuzzy control model corresponding to the disturbance state based on the judgment result; A model inference module, used for inputting the operating data into a target fuzzy control model corresponding to the disturbance state, and performing inference calculation based on the operating data using the target fuzzy control model to obtain a target mechanical frequency output by the target fuzzy control model; The mechanical control module is used to control the motor disposed in the large rotating machine to operate according to the target mechanical frequency.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the control method for large rotating machinery as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the control method for a large rotating machine as claimed in any one of claims 1 to 7 are executed.

Images