Horizontal centrifuge real-time PID gain optimization control method based on improved adaptive genetic algorithm

Abstract

The invention discloses a horizontal centrifuge real-time PID gain optimization control method based on an improved adaptive genetic algorithm. The method comprises the steps that a chromosome population is defined based on PID gain of a horizontal centrifuge; constructing a first fitness function based on the first index of the horizontal centrifuge, and calculating the fitness of individuals in the chromosome population; and screening parent individuals by using a champion for genetic generation of offspring, obtaining the crossover rate and mutation rate of the individuals based on the fitness of the individuals in the genetic process, repeating the genetic process until convergence, and applying the PID gain corresponding to the current optimal individual to control the horizontal centrifuge. In the process, the parent is screened, and the genetic crossover rate and the mutation rate are adaptively adjusted, so that the PID parameters are converged more quickly, and the robustness is high.

Description

Technical Field

[0001] The application relates to the field of PID controllers, and in particular to a real-time PID gain optimization control algorithm for horizontal centrifuges based on an improved adaptive genetic algorithm. Background Technology

[0002] The PID (Proportional-Integral-Derivative) controller is one of the most classic control algorithms in industrial control. The PID controller calculates the control output based on the system's current error (the deviation between the setpoint and the actual value), its historical changes (integral), and its future trend (derivative). In other words, it achieves stable, fast, and precise control by adjusting the system's error signal. It has a simple structure and strong robustness, and is widely used in temperature control, motor speed regulation, robotics, chemical processes, and other fields.

[0003] Traditional PID controllers often rely on empirical tuning or single-shot tuning methods such as the Ziegler-Nichols and Cohen-Kuhn approaches for their gain parameters, making it difficult to balance speed and steady-state accuracy. For example, horizontal centrifuges are widely used in sludge treatment, wastewater purification, chemical, and mining industries due to their high-efficiency solid-liquid separation capabilities. However, horizontal centrifuges suffer from slow response, large overshoot, and long convergence times during startup and load fluctuations when adjusting speed.

[0004] In recent years, genetic algorithms (GA) have been used for PID parameter optimization. They search for optimal PID parameters, such as proportional gain, integral time, and derivative time, in the parameter space by simulating natural selection and genetic mechanisms (e.g., selection, crossover, mutation) to improve the dynamic performance of control systems, such as overshoot, settling time, and steady-state error. However, standard genetic algorithms suffer from slow convergence and a tendency to get trapped in local optima. Summary of the Invention

[0005] This application provides a real-time PID gain optimization control method for horizontal centrifuges based on an improved adaptive genetic algorithm, which solves the defects of slow convergence speed and easy getting trapped in local optima in the standard genetic algorithm.

[0006] According to one aspect of this application, a real-time PID gain optimization control method for a horizontal centrifuge based on an improved adaptive genetic algorithm is provided, comprising:

[0007] Define chromosome population based on PID gain of horizontal centrifuge;

[0008] The first fitness function is constructed based on the first index of the horizontal centrifuge, and the fitness of individuals in the chromosome population is calculated.

[0009] The process involves using a tournament to select parent individuals for genetic generation of offspring. During the genetic process, the crossover rate and mutation rate of individuals are obtained based on their fitness. This genetic process is repeated until convergence, and the PID gain corresponding to the best individual in the current generation is applied to control a horizontal centrifuge.

[0010] Furthermore, the method of using a tournament to select parent individuals for genetic generation of offspring includes:

[0011] A tournament group is formed by randomly selecting N individuals from the contemporary era.

[0012] The individual with the lowest fitness in the tournament group was selected as the parent.

[0013] Repeat the above process until the parent generation reaches the population size.

[0014] Furthermore, the first fitness function includes several types based on the operating state of the horizontal centrifuge, among which:

[0015] When the horizontal centrifuge is in the startup phase, the first fitness function is: Where ω1 and ω2 are weights, and ω1 is greater than ω2, T1 is the time taken to reach the target speed, and Mp is the overshoot;

[0016] When the horizontal centrifuge is in a load fluctuation phase, the first fitness function is: Where ω2 and ω4 are weights, and ω3 is greater than ω4, and the magnitude of ω3 is positively correlated with the frequency of load fluctuations, Δ is the variance of the rotating beam, and T2 is the time taken for the load fluctuation to recover;

[0017] When the horizontal centrifuge is in a stable operating phase, the first fitness function fitness(x) = |ω set -ω(x)|, where ω set Let ω(x) be the target rotational speed, ω(x) be the actual rotational speed, and x be an individual in the population.

[0018] Furthermore, it also includes constructing a digital twin model of a horizontal centrifuge and defining the chromosome population of the digital twin model based on the PID gain of the digital twin model;

[0019] Based on the second index, a second fitness function is constructed for individuals in the chromosome population of the digital twin model, and the fitness of individuals in the chromosome population of the digital twin model is calculated. The second index includes the normalized value D of the maximum rotational speed deviation when a load exceeding the threshold is suddenly applied, the normalized value M of the parameter drift after 24 hours of continuous operation, and the normalized value A of the vibration amplitude growth rate under bearing wear conditions. The second fitness function is fitness(x) = ω5D + ω6M + ω7A, where ω5, ω6, and ω7 are weights.

[0020] The genetic process of the horizontal centrifuge and its digital twin model is executed synchronously. Parent individuals are screened using a tournament to obtain two corresponding parent chromosome populations for the horizontal centrifuge and the digital twin model, respectively. Several parent individuals from the digital twin model are selected and added to the parent chromosome population of the horizontal centrifuge to obtain the current parent chromosome population of the horizontal centrifuge. Offspring are generated based on the current parent chromosome population.

[0021] Furthermore, the step of selecting several parent individuals from the digital twin model and adding them to the parent chromosome population of the horizontal centrifuge to obtain the current generation parent chromosome population includes:

[0022] When the horizontal centrifuge is in a stable operating phase, select 5%-10% of the parent individuals in the digital twin model to be added to the parent chromosome population of the horizontal centrifuge.

[0023] When the horizontal centrifuge is in a load fluctuation phase, select 20%-30% of the parent chromosome population of the horizontal centrifuge from the digital twin model, and replace the corresponding number of individuals with low fitness in the parent chromosome population of the horizontal centrifuge.

[0024] Furthermore, the crossover rate P of an individual is obtained based on its fitness. c (x) is as follows:

[0025]

[0026] Where fitness(x) is the fitness function, fitness min For minimum fitness, fitness avg For average fitness, PC max PC is the upper limit of the cross rate. min This is the lower bound of the crossover rate.

[0027] Furthermore, the variation rate P of an individual is obtained based on its fitness. m (x) is as follows:

[0028]

[0029] Among them, fitness max For maximum fitness, PM max PM is the upper limit of the mutation rate. min This is the lower limit of the variation rate.

[0030] A second aspect of the present invention provides a real-time PID gain optimization control device for a horizontal centrifuge based on an improved adaptive genetic algorithm, comprising:

[0031] The population definition module is used to define chromosome populations based on the PID gain of a horizontal centrifuge.

[0032] The fitness calculation module is used to construct a first fitness function based on the first index of the horizontal centrifuge and to calculate the fitness of individuals in the chromosome population.

[0033] The genetic module is used to select parent individuals through a tournament to generate offspring. During the genetic process, the crossover rate and mutation rate of individuals are obtained based on their fitness. The genetic process is repeated until convergence, and the PID gain corresponding to the best individual in the current generation is applied to control the horizontal centrifuge.

[0034] A third aspect of the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, the electronic device performs the method as described in the first aspect of the present invention.

[0035] A fourth aspect of the present invention provides a computer-readable storage medium for storing a computer program that, when run on a computer, causes the computer to perform the method described in the first aspect of the present invention.

[0036] In this embodiment, a real-time PID gain optimization control method for a horizontal centrifuge based on an improved adaptive genetic algorithm is employed. The method includes: defining a chromosome population based on the PID gain of the horizontal centrifuge; constructing a first fitness function based on a first index of the horizontal centrifuge and calculating the fitness of individuals in the chromosome population; using a tournament to select parent individuals for genetic generation of offspring, wherein the crossover rate and mutation rate of individuals are obtained based on their fitness during the genetic process, and the above genetic process is repeated until convergence. The PID gain corresponding to the best individual in the current generation is then applied to control the horizontal centrifuge. In this process, the selection of parent individuals and the adaptive adjustment of the genetic crossover rate and mutation rate enable the PID parameters to converge faster and exhibit higher robustness. Attached Figure Description

[0037] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0038] Figure 1 This is a flowchart of a real-time PID gain optimization control method for a horizontal centrifuge based on an improved adaptive genetic algorithm, according to an embodiment of this application.

[0039] Figure 2 This is a schematic diagram of the parent selection method according to the embodiments of this application.

[0040] Figure 3 This is a schematic diagram of the convergence process formed by regulating the speed of a horizontal centrifuge using the traditional PID algorithm, the traditional genetic algorithm, and the method in this application, according to the embodiments of this application.

[0041] Figure 4 This is a schematic diagram illustrating the real-time changes of the proportional, integral, and derivative parameters during PID optimization according to the method in the embodiments of this application.

[0042] Figure 5 This is a schematic diagram of a real-time PID gain optimization control device for a horizontal centrifuge based on an improved adaptive genetic algorithm, according to an embodiment of this application.

[0043] Figure 6 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0044] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0045] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0046] Embodiments of the present invention provide a real-time PID gain optimization control method for horizontal centrifuges based on an improved adaptive genetic algorithm, used to improve the PID control parameters for horizontal centrifuges, such as... Figure 1 The diagram shown is a flowchart illustrating a real-time PID gain optimization control method for a horizontal centrifuge based on an improved adaptive genetic algorithm, as proposed in this application. Figure 1 It can be seen that the real-time PID gain optimization control method for horizontal centrifuges based on an improved adaptive genetic algorithm includes the following steps:

[0047] Step S102: Define a chromosome population based on the PID gain of a horizontal centrifuge, where each individual is (Kp, Ki, Kd) and the population size is N, where Kp is the proportional coefficient, Ki is the integral time, and Kd is the derivative time.

[0048] Step S104: Construct a first fitness function based on the first index of the horizontal centrifuge, and calculate the fitness of individuals in the chromosome population. The first fitness function is a standard for evaluating the quality of parameters; the lower the first fitness of an individual, the better its parameters.

[0049] Since horizontal centrifuges may operate under various conditions, adjusting the first fitness function according to these conditions allows for faster evolution of individuals suited to the current conditions. In some preferred embodiments, the first fitness function includes several variations based on the operating state of the horizontal centrifuge, among which:

[0050] When a horizontal centrifuge is in the startup phase, it is desirable to quickly reach the target speed, but overshoot must be avoided. Therefore, the first fitness function is: Where ω1 and ω2 are weights, and since this stage focuses on the start-up time, ω1 is set to be greater than ω2. T1 is the time taken to reach the target speed, and Mp is the overshoot.

[0051] When a horizontal centrifuge is in a load fluctuation phase, it is desirable to adapt to the fluctuations and stabilize as quickly as possible. Therefore, the first fitness function is: Where ω2 and ω4 are weights, and since stability is of greater concern at this time, ω3 is set to be greater than ω4, and the magnitude of ω3 is positively correlated with the frequency of load fluctuations. Δ is the variance of the rotating beam, and T2 is the time taken for the load fluctuation to recover.

[0052] When the horizontal centrifuge is in a stable operating phase, the focus is more on whether the rotational speed meets the target. Therefore, the first fitness function is set as fitness(x) = |ω set -ω(x)|, where ω set Let ω(x) be the target rotational speed, ω(x) be the actual rotational speed, and x be an individual in the population.

[0053] By adjusting the first fitness function in real time under different working conditions, the genetic mutation is made to fit the current working scenario.

[0054] Step S106: Use the tournament to screen parent individuals for genetic generation of offspring. During the genetic process, the crossover rate and mutation rate of individuals are obtained based on their fitness. Repeat the above genetic process until convergence. Apply the PID gain corresponding to the best individual in the current generation to control the horizontal centrifuge.

[0055] By using the above method, the PID control parameters are adjusted under the action of the optimized genetic algorithm, which makes the PID control converge faster and can make adaptive adjustments according to the working conditions, resulting in stronger robustness and higher adaptability.

[0056] In some preferred embodiments, such as Figure 2 As shown, the method of using a tournament to select parent individuals for genetic generation of offspring includes the following steps:

[0057] Step S1061: Randomly select N individuals from the current generation to form a tournament group;

[0058] Step S1062: Select the individual with the lowest fitness in the tournament group as the parent generation;

[0059] Step S1063: Repeat the above process until the parent generation reaches the population size.

[0060] For example, in some embodiments, N is arbitrarily selected as 3 each time to obtain 1 parent generation. By repeating the above selection of parent generation multiple times, multiple parent generations can be obtained.

[0061] For example, in some embodiments, N is arbitrarily selected as 8 each time to obtain 1 parent generation. By repeating the above selection of parent generation multiple times, multiple parent generations can be obtained.

[0062] For example, in some embodiments, N is arbitrarily selected each time, which is any value between 3 and 8, to obtain 1 parent generation. By repeating the above selection of parent generation multiple times, multiple parent generations can be obtained.

[0063] In some preferred embodiments, in order to enable better control of the horizontal centrifuge under extreme conditions, the method further includes constructing a digital twin model of the horizontal centrifuge, defining the chromosome population of the digital twin model based on the PID gain of the digital twin model, simulating the operation under extreme conditions through the digital twin model, and running a genetic algorithm under extreme conditions to obtain individuals that can adapt to extreme conditions.

[0064] Specifically, in these embodiments, a second fitness function for individuals in the chromosome population of the digital twin model is constructed based on a second index, and the fitness of individuals in the chromosome population of the digital twin model is calculated. The second index represents several indices under extreme working conditions, including the normalized value D of the maximum speed deviation when a load exceeding a threshold is suddenly applied, the normalized value M of the parameter drift after 24 hours of continuous operation, and the normalized value A of the vibration amplitude growth rate under bearing wear conditions. The second fitness function is fitness(x) = ω5D + ω6M + ω7A, where ω5, ω6, and ω7 are weights.

[0065] In some optional embodiments, ω5, ω6 and ω7 can be dynamically adjusted. For example, when the time of the sudden increase in load exceeding the threshold is within the set time threshold range, the weight of ω5 is greater than that of ω6 and ω7. When the time threshold range is exceeded, the weight of ω5 gradually decreases as time is delayed. Here, the load exceeding the threshold can be, for example, 150% or more, 200% or more, or 300% or more of the rated load.

[0066] For example, after the digital twin model has been running continuously for more than 24 hours, the weight of ω6 is increased, and ω6 gradually increases as time goes on.

[0067] For example, when the digital twin model changes from the original unworn bearing condition to the worn bearing condition, the weight of the original ω7 is increased.

[0068] It should be noted that in the above embodiments, the three weights of the digital twin model are set with upper and lower limits, and all the adjustments mentioned above are made within the range of the upper and lower limits of the weights.

[0069] In this embodiment, the genetic process of the horizontal centrifuge and its digital twin model is executed simultaneously. The horizontal centrifuge and the digital twin model are respectively screened using a tournament to obtain two sets of parent chromosome populations. Several parent individuals from the digital twin model are selected and added to the parent chromosome population of the horizontal centrifuge to obtain the current generation of parent chromosome population of the horizontal centrifuge. Offspring are generated based on the current generation of parent chromosome population.

[0070] In some preferred embodiments, the step of selecting several parent individuals from the digital twin model and adding them to the parent chromosome population of the horizontal centrifuge to obtain the current parent chromosome population includes:

[0071] When the horizontal centrifuge is in a stable operating phase, select 5%-10% of the parent individuals in the digital twin model to be added to the parent chromosome population of the horizontal centrifuge.

[0072] When the horizontal centrifuge is in a load fluctuation phase, select 20%-30% of the parent chromosome population of the horizontal centrifuge from the digital twin model, and replace the corresponding number of individuals with low fitness in the parent chromosome population of the horizontal centrifuge.

[0073] By selecting different numbers of parent individuals for different operating conditions of horizontal centrifuges, the parent individuals of the horizontal centrifuges can be made more adaptable to different operating conditions and can complete iterations more quickly.

[0074] In some preferred embodiments, the crossover rate P of individuals is obtained based on their fitness. c (x) is as follows:

[0075]

[0076] Where fitness(x) is the fitness function, fitness min For minimum fitness, fitness avg For average fitness, PC max PC is the upper limit of the cross rate. min This is the lower bound of the crossover rate.

[0077] In some preferred embodiments, the variation rate P of an individual is obtained based on its fitness. m (x) is as follows:

[0078]

[0079] Among them, fitness max For maximum fitness, PM max PM is the upper limit of the mutation rate. min This is the lower limit of the variation rate.

[0080] Through the adaptive adjustment of crossover and mutation rates, for the crossover rate, when the fitness of an individual far exceeds the average fitness, a higher crossover probability is assigned to encourage the recombination of superior genes, enhance the global search capability, and prevent individuals from converging towards inferior directions; when the fitness of an individual is close to the average level, a lower crossover probability is assigned to protect the superior genes of average individuals.

[0081] Regarding the mutation rate, when an individual's fitness is low, it means that the individual has superior genes, so a low mutation rate is assigned to avoid destroying the superior genes that have been obtained; when an individual's fitness is high, a high mutation rate is assigned to allow for a wide range of exploration and to escape the local region of inferior solutions.

[0082] Figure 3 The process of controlling the time-varying PID parameters using three different methods is illustrated. Figure 3 As can be seen, the method of this embodiment can respond rapidly in about 0.2s, rising to about 4200rpm, and converge to close to 4000rpm in about 0.3s. The magnified graph shows that it fully converges to 4000rpm around 3s. The standard genetic algorithm rises to about 4300rpm in about 0.75s, then gradually begins to converge, approaching 4000rpm after 2s, and the magnified graph shows that it only fully converges to 4000rpm after 3.5s. The traditional PID rises to about 5100rpm in about 0.5s and falls to 3700rpm in about 0.9s, with a large overshoot and significant fluctuations. The magnified graph shows that it only approaches 4000rpm after 4s, indicating slow convergence. In summary, compared to the other two methods, the method of this invention has a faster response speed, smaller overshoot, and faster convergence, further demonstrating the effectiveness of the method and its excellent adjustment effect on centrifuge speed.

[0083] To further demonstrate the real-time adjustability of the proportional, integral, and differential parameters of the method of the present invention, Figure 4 The real-time changes of the three parameters are shown throughout the simulation. From Figure 4 The changes of the three parameters over time can be observed, ensuring that the proportional, integral, and derivative parameters used in each feedback loop are optimal for the current time.

[0084] Based on the same inventive concept, embodiments of this application also provide a real-time PID gain optimization control device for a horizontal centrifuge based on an improved adaptive genetic algorithm, such as... Figure 5 As shown, the device includes:

[0085] The population definition module is used to define chromosome populations based on the PID gain of a horizontal centrifuge.

[0086] The fitness calculation module is used to construct a first fitness function based on the first index of the horizontal centrifuge and to calculate the fitness of individuals in the chromosome population.

[0087] The genetic module is used to select parent individuals through a tournament to generate offspring through genetic inheritance. During the inheritance process, the crossover rate and mutation rate of individuals are obtained based on their fitness. This genetic process is repeated until convergence, and the PI corresponding to the current best individual is determined. It should be noted that although several units or sub-units of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0088] Based on the same inventive concept as the above method embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it enables the electronic device to implement the control method described in the above embodiments.

[0089] In one embodiment, the electronic device may be a server, and in this embodiment, the structure of the electronic device may be as follows: Figure 6 As shown, it includes a memory 2001, a communication module 2003, and one or more processors 2002.

[0090] The memory 2001 is used to store computer programs executed by the processor 2002. The memory 2001 may mainly include a program storage area and a data storage area. The program storage area may store the operating system and programs required to run instant messaging functions, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0091] Memory 2001 may be volatile memory, such as random-access memory (RAM); memory 2001 may also be non-volatile memory, such as read-only memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory 2001 may be any other medium capable of carrying or storing a desired computer program having the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 2001 may be a combination of the above-mentioned memories.

[0092] Processor 2002 may include one or more central processing units (CPUs) or digital processing units, etc. Processor 2002 is used to implement the above-mentioned audio data processing method when calling computer programs stored in memory 2001.

[0093] The communication module 2003 is used to communicate with terminal devices and other servers.

[0094] This application embodiment does not limit the specific connection medium between the memory 2001, communication module 2003, and processor 2002. This application embodiment... Figure 6 The memory 2001 and the processor 2002 are connected via a bus 2004, which is in... Figure 6 The connections between other components are illustrated with arrows and are for illustrative purposes only, not as limiting information. The Bus 2004 can be divided into address bus, data bus, control bus, etc. For ease of description, Figure 6 The text uses only one arrow to describe it, but does not indicate that there is only one bus or one type of bus.

[0095] Based on the same inventive concept as the above-described method embodiments, embodiments of the present invention also provide a computer-readable storage medium for storing a computer program. When the computer program is run on a computer, it enables the electronic device to implement the control method described in the above embodiments. The computer-readable storage medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0096] Based on the same inventive concept as the above-described method embodiments, embodiments of the present invention also provide a computer program product, which includes a computer program that, when run on an electronic device, causes the electronic device to perform the steps of the control methods described above according to various exemplary embodiments of this application. The program product may take the form of any combination of one or more readable media. These computer program commands can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the commands executed by the processor of the computer or other programmable data processing device generate a process for implementing... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

Claims

1. A real-time PID gain optimization control method for a horizontal centrifuge based on an improved adaptive genetic algorithm, characterized in that, Define chromosome population based on PID gain of horizontal centrifuge; The first fitness function is constructed based on the first index of the horizontal centrifuge, and the fitness of individuals in the chromosome population is calculated. The process involves using a tournament to select parent individuals for genetic generation of offspring. During the genetic process, the crossover rate and mutation rate of individuals are obtained based on their fitness. This genetic process is repeated until convergence, and the PID gain corresponding to the best individual in the current generation is applied to control a horizontal centrifuge.

2. The method according to claim 1, characterized in that, The method of using a tournament to select parent individuals for genetic generation of offspring includes: A tournament group is formed by randomly selecting N individuals from the contemporary era. The individual with the lowest fitness in the tournament group was selected as the parent. Repeat the above process until the parent generation reaches the population size.

3. The method according to any one of claims 1 to 2, characterized in that, The first fitness function includes several types based on the operating state of the horizontal centrifuge, among which: When the horizontal centrifuge is in the startup phase, the first fitness function is: Where ω1 and ω2 are weights, and ω1 is greater than ω2, T1 is the time taken to reach the target speed, and Mp is the overshoot; When the horizontal centrifuge is in a load fluctuation phase, the first fitness function is: Where ω2 and ω4 are weights, and ω3 is greater than ω4, and the magnitude of ω3 is positively correlated with the frequency of load fluctuations, Δ is the variance of the rotating beam, and T2 is the time taken for the load fluctuation to recover; When the horizontal centrifuge is in a stable operating phase, the first fitness function fitness(x) = |ω set -ω(x)|, where ω set Let ω(x) be the target rotational speed, ω(x) be the actual rotational speed, and x be an individual in the population.

4. The method according to claim 3, characterized in that, It also includes constructing a digital twin model of a horizontal centrifuge and defining the chromosome population of the digital twin model based on the PID gain of the digital twin model; Based on the second index, a second fitness function is constructed for individuals in the chromosome population of the digital twin model, and the fitness of individuals in the chromosome population of the digital twin model is calculated. The second index includes the normalized value D of the maximum rotational speed deviation when a load exceeding the threshold is suddenly applied, the normalized value M of the parameter drift after 24 hours of continuous operation, and the normalized value A of the vibration amplitude growth rate under bearing wear conditions. The second fitness function is fitness(x) = ω5D + ω6M + ω7A, where ω5, ω6, and ω7 are weights. The genetic process of the horizontal centrifuge and its digital twin model is executed synchronously. Parent individuals are screened using a tournament to obtain two corresponding parent chromosome populations for the horizontal centrifuge and the digital twin model, respectively. Several parent individuals from the digital twin model are selected and added to the parent chromosome population of the horizontal centrifuge to obtain the current parent chromosome population of the horizontal centrifuge. Offspring are generated based on the current parent chromosome population.

5. The method according to claim 4, characterized in that, The process of selecting several parent individuals from the digital twin model and adding them to the parent chromosome population of the horizontal centrifuge to obtain the current generation of parent chromosomes includes: When the horizontal centrifuge is in a stable operating phase, select 5%-10% of the parent individuals in the digital twin model to be added to the parent chromosome population of the horizontal centrifuge. When the horizontal centrifuge is in a load fluctuation phase, select 20%-30% of the parent chromosome population of the horizontal centrifuge from the digital twin model, and replace the corresponding number of individuals with low fitness in the parent chromosome population of the horizontal centrifuge.

6. The method according to claim 5, characterized in that, The crossover rate P of individuals is obtained based on their fitness. c (x) is as follows: Where fitness(x) is the fitness function, fitness min For minimum fitness, fitness avg For average fitness, PC max PC is the upper limit of the cross rate. min This is the lower bound of the crossover rate.

7. The method according to claim 5, characterized in that, The individual's variability P is obtained based on the individual's fitness. m (x) is as follows: Among them, fitness max For maximum fitness, PM max PM is the upper limit of the mutation rate. min This is the lower limit of the variation rate.

8. A real-time PID gain optimization control device for a horizontal centrifuge based on an improved adaptive genetic algorithm, characterized in that, include: The population definition module is used to define chromosome populations based on the PID gain of a horizontal centrifuge. The fitness calculation module is used to construct a first fitness function based on the first index of the horizontal centrifuge and to calculate the fitness of individuals in the chromosome population. The genetic module is used to select parent individuals through a tournament to generate offspring. During the genetic process, the crossover rate and mutation rate of individuals are obtained based on their fitness. The genetic process is repeated until convergence, and the PID gain corresponding to the best individual in the current generation is applied to control the horizontal centrifuge.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the electronic device to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 7.

Images