Optimal control method for free piston internal combustion generator power system based on fuzzy PI control

Abstract

The application discloses a kind of based on fuzzy PI control's free piston internal combustion generator power system optimization control method, comprising: constructing the dynamics equation under the working state of FPEG system, and the dynamics model is obtained according to the dynamics equation;Establish fuzzy PI controller, the fuzzy PI controller includes fuzzy controller and PI controller, the fuzzy controller is used to adjust the proportional parameter K p And integral parameter K i Of PI controller;Fuzzy PI controller is applied to dynamics model, forms double closed loop feedback, and the piston stroke of FPEG system is controlled.Using genetic algorithm to optimize the parameters of the fuzzy PI controller.The fuzzy PI control formed by combining fuzzy control algorithm and PI control method in the application controls the piston stroke of FPEG system, and has the advantages of improving response speed, reducing tracking error and the like.Again, the parameters of the fuzzy PI controller designed under the sinusoidal input reference signal are optimized using genetic algorithm, to realize the control optimization of FPEG system.

Description

Technical Field

[0001] This invention relates to the field of free piston internal combustion generator control technology, and specifically to an optimized control method for a free piston internal combustion generator power system based on fuzzy PI control. Background Technology

[0002] The Free Piston Engine Linear Generator (FPEG) is a novel power unit. Compared to traditional reciprocating piston internal combustion engines, the FPEG couples the free piston internal combustion engine with a linear generator, with the piston connected to the mover of the linear generator, forming the only significant moving part. The FPEG features high energy conversion efficiency, high power density, variable compression ratio operation, and less friction and wear, while also boasting a compact structure, making it a research hotspot in recent years for novel power machinery. Piston motion control is a crucial aspect of the FPEG's operation. Due to its demanding requirements for speed and stability, traditional PID control methods are generally considered. While PID control structures are easy to implement, parameter tuning is cumbersome, often relying on manual adjustments based on engineer experience. This constant adjustment leads to significant errors, making it difficult to achieve ideal performance for the entire control system. Therefore, although PID control is the most commonly used control scheme in industrial production processes, it cannot meet the control requirements of some more complex systems.

[0003] With the development of various industries, especially in fields such as electric vehicle power and marine power, the efficiency requirements for power units are becoming increasingly higher, which places higher demands on the piston control of FPEG. Therefore, developing a new type of FPEG piston motion controller has significant application value in order to improve control accuracy. Actual FPEG systems are large and complex systems with many time-varying uncertainties and nonlinearities, making precise modeling difficult. Therefore, there is an urgent need to find a control method that does not require a precise mathematical model. Fuzzy control is an intelligent control method that mimics the fuzzy reasoning and decision-making process of humans. Based on fuzzy set theory, fuzzy linguistic variables, and fuzzy logic reasoning, fuzzy control does not require a precise mathematical model of the controlled object. Instead, it uses control rules summarized from a large amount of actual operational data and expresses the control strategy in natural language. This method first compiles the experience of operators or experts into fuzzy rules, then fuzzifies the real-time signals from sensors, uses the fuzzified signals as input to the fuzzy rules to complete fuzzy reasoning, and adds the output obtained after reasoning to the actuator. For complex control systems where the controlled object exhibits characteristics such as hysteresis, nonlinearity, and coupling, it is difficult to obtain the extensive system knowledge required for precise control. In such cases, fuzzy control can be employed to meet the demand for precise control of these complex systems. Given the advantages of fuzzy control, such as not requiring precise mathematical models and strong robustness, a fuzzy PID controller can be used to control the system. The fuzzy PID controller, based on the traditional PID controller, uses a specific fuzzy algorithm to self-tune and optimize the PID parameters in real time, overcoming the limitation of traditional PID controllers that cannot adjust parameters in real time, thereby improving control performance.

[0004] To meet the more precise output requirements of the power system, an optimization algorithm is used to optimize the parameters of the fuzzy PID controller. The Genetic Algorithm (GA) is a search algorithm based on natural selection and population genetics mechanisms. It draws on Darwin's theory of evolution and Mendel's theory of heredity, simulating the phenomena of reproduction, hybridization, and mutation in natural selection and heredity. It adopts the principle of "survival of the fittest," eliminating "bad" individuals in each generation and retaining "good" individuals. Essentially, it is an efficient, parallel, and global search method.

[0005] Genetic operations essentially simulate the operations of biological genes. Therefore, when using genetic algorithms to solve problems, the variables must first be encoded. Each potential solution to the problem is encoded as a "chromosome," i.e., an individual. Several individuals constitute a population (all possible solutions). At the start of the genetic algorithm, some individuals (i.e., initial solutions) are randomly generated. Each individual is evaluated according to a predetermined objective function, and a fitness value is assigned. Based on this fitness value, a selection operation is performed. Superior individuals are selected from the population, while inferior individuals are eliminated. Individuals with higher fitness are more likely to be selected, reflecting the principle of "survival of the fittest." The selected individuals are placed in a pairing pool. Two individuals are randomly selected from the pairing pool, and through crossover and mutation operations, a new generation (offspring) is generated. Offspring individuals inherit the superior genes of their parents, and therefore outperform the previous generation, i.e., evolution. This process is repeated until the convergence condition is met. In summary, selection achieves the effect of "survival of the fittest," crossover ensures the stability of the population, and evolves towards the optimal solution. Selection and crossover basically complete most of the search functions of the genetic algorithm, while mutation ensures the diversity of the population, avoids local convergence that may be caused by crossover, and increases the ability of the genetic algorithm to find the optimal solution.

[0006] Currently, when using the PID control method to control the piston in the FPEG system, there are problems such as large tracking error, slow tracking speed, and complicated parameter tuning. Therefore, it is urgent to optimize the PID control method to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to propose a fuzzy PID control method that combines fuzzy control algorithm and PID control method to control the piston stroke of FPEG system, which has the advantages of improving response speed and reducing tracking error.

[0008] To achieve the above objectives, this application proposes an optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control, comprising:

[0009] Construct the dynamic equations of the FPEG system under its operating conditions, and obtain the dynamic model based on the dynamic equations.

[0010] A fuzzy PI controller is established, comprising a fuzzy controller and a PI controller, wherein the fuzzy controller is used to adjust the proportional parameter K of the PI controller. p and integration parameter K i ;

[0011] A fuzzy PI controller is applied to the dynamic model to form a double closed-loop feedback, which controls the piston stroke of the FPEG system.

[0012] Furthermore, it also includes: using a genetic algorithm to optimize the parameters of the fuzzy PI controller.

[0013] Furthermore, when the FPEG system is operating in electric motor mode, the piston reciprocates linearly within the cylinder, and the control objective of the electric motor is to make the piston's trajectory track a sinusoidal signal of a certain frequency.

[0014] Furthermore, considering the piston motion components of the FPEG system as a mass-spring-damped system, its overall dynamic equation is: By controlling the current I q To control F e This controls the FPEG system to make the piston motion a sinusoidal trajectory, where m is the mass of the moving component (kg), x is the piston trajectory, and F... l F represents the gas pressure (N) from the left cylinder. r F represents the gas pressure (N) from the right cylinder. e For electromagnetic resistance (N), F f This is the sum of the frictional forces acting on the piston and the moving part assembly.

[0015] Furthermore, by leveraging the input-output relationship between the fuzzy controller and the PI algorithm, the fuzzy controller and the PI controller are combined to form a fuzzy PI controller.

[0016] Furthermore, the input signal of the fuzzy controller is the positional deviation e of the piston motion assembly stroke. x and the rate of change of position deviation (ec) x The output signal is the parameter adjustment amount ΔK of the outer loop PI controller. p ΔK i The quantization factor of the fuzzy controller is K. ex K ecx The scaling factors are K1 and K2.

[0017] Furthermore, the fuzzy subset of the input variables of the fuzzy controller is {negative large NB, negative medium NM, negative small NS, zero ZE, positive small PS, positive medium PM, positive large PB}, and the membership functions of the input variables are all triangular, and the membership functions of the output variables are all triangular in the middle and Gaussian at both ends; the fuzzy controller adopts the Mamdani inference method, sets control rules, and expresses the rule table in the form of if-then, with a total of 49 fuzzy rules.

[0018] Furthermore, the dynamic model applies a fuzzy PI controller to the outer loop and a PI controller to the inner loop, forming a dual closed-loop feedback control; the fuzzy PI controller outputs a speed control command for the piston motion assembly, and the PI controller outputs a current I for the piston motion assembly. q instruction.

[0019] As a further step, a GA genetic algorithm is used to optimize the quantization factor and scaling factor of the fuzzy PI controller in the outer loop.

[0020] Compared with existing technologies, the technical solution adopted in this invention has the following advantages: This invention combines fuzzy control algorithms and PI control methods to form fuzzy PI control, which controls the piston stroke of the FPEG system. This improves response speed and reduces tracking errors, thereby achieving better trajectory control. Furthermore, a genetic algorithm is used to optimize the parameters of the fuzzy PI controller designed under a sinusoidal input reference signal, thus optimizing the control of the FPEG system. The method proposed in this invention significantly improves the dynamic and steady-state performance of the piston stroke in the power system, exhibiting strong robustness and wide applicability. Attached Figure Description

[0021] Figure 1 The flowchart shows the implementation of the optimized control method for the free piston internal combustion generator power system based on fuzzy PI control in the embodiment.

[0022] Figure 2 This is a schematic diagram of the dynamic model structure in the embodiment;

[0023] Figure 3 This is a structural diagram of the fuzzy PI controller in the embodiment;

[0024] Figure 4 This is the rule control diagram of the fuzzy controller in the embodiment;

[0025] Figure 5 This is a simulation model diagram of the fuzzy PI controller in the embodiment;

[0026] Figure 6 This is a schematic diagram of the FPEG system control principle in the embodiment;

[0027] Figure 7 This is a simulation diagram of the FPEG system control in the embodiment;

[0028] Figure 8 The flowchart for the GA genetic algorithm optimization in this embodiment is shown.

[0029] Figure 9 This is a simulation diagram of the GA genetic algorithm optimization in the embodiment;

[0030] Figure 10 This is a comparison chart of the control indicators of the FPEG system in the embodiments;

[0031] Figure 11-13 This is a comparison chart of the control effects of the FPEG system in the embodiments. Specific implementation methods

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit the application; that is, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0033] Example 1

[0034] like Figure 1 As shown, this invention provides an optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control. The method includes the following steps:

[0035] S1. Construct the dynamic equations of the FPEG system under its working state, and obtain the dynamic model based on the dynamic equations;

[0036] In this embodiment, the dual-piston, dual-cylinder FPEG system is the main research object. The mover of the linear motor is connected to the piston and placed between the two engine cylinders, forming the only main moving part in this structure. The expansion and power stroke is completed alternately in the two cylinders, and the piston moving assembly is driven by the combustion explosion pressure to overcome the gas compression force in the other cylinder. When the FPEG system is operating in electric motor mode, the piston performs reciprocating linear motion in the cylinder. The control objective of the electric motor is to make the piston motion trajectory track a sinusoidal signal of a certain frequency.

[0037] Specifically, considering the piston motion components of the FPEG system as a mass-spring-damped system, its overall dynamic equation is: In three-phase motor control, to reduce control complexity, the three-phase system is transformed from a natural coordinate system to a stationary coordinate system, and further to a synchronous moving coordinate system. Ultimately, the motor current is characterized by the q-axis and d-axis currents. A vector control approach is adopted, always keeping the d-axis current at zero, and controlling the motor torque (thrust) through the q-axis current. When the FPEG system operates in motor mode, That is, by controlling the current I q To control F e For a typical experimental motor, kq = 78.9 N / A. In motor mode, friction can be simplified to a damping force related to the speed of the moving component, and the sum of the two can be linearized as follows: Therefore, the dynamic equations of this system can be simplified to: The change in gas pressure within the cylinder is mainly driven by two factors: the change in working volume and the heat released during combustion. Using Taylor expansion, equivalent linearization is performed, followed by Fourier series expansion, finally yielding the kinetic equations for the FPEG system:

[0038]

[0039] Based on the current I mentioned above q To establish a dynamic model of the FPEG system, the relationship between position x and velocity v is used. Figure 2 As shown. The input to the FPEG system dynamics model is the current command I. q The position x and velocity v of the piston motion component are used to output the acceleration of the piston motion component.

[0040] S2. Establish a fuzzy PI controller, which includes a fuzzy controller and a PI controller. The fuzzy controller is used to adjust the proportional parameter K of the PI controller. p and integration parameter K i ;

[0041] Specifically, the control objective of the FPEG system is analyzed, and fuzzy control is combined with PI control to establish a fuzzy PI controller. The fuzzy PI controller includes a fuzzy controller and a PI controller, wherein the fuzzy controller is used to adjust the control parameter K of the PI controller. p ,K i .

[0042] It should be noted that: such as Figure 3 As shown, the fuzzy controller structure is a two-dimensional structure, using the error e0 changing at different times and the error change ec0 as its input values. In the system of this invention, the input signal of the fuzzy controller is the position deviation e of the piston motion component's stroke. x and the rate of change of position deviation (ec) x The output signal is the parameter adjustment amount ΔK of the outer loop PI controller. p ΔK i The obtained ΔK p ΔK i Adding the parameters to the original PI controller parameters yields the real-time PI control parameters. The fuzzy controller comprises four parts: a fuzzification part, which converts clear input variables into fuzzy quantities; a knowledge base part, containing membership functions, quantization factors, scaling factors, and a fuzzy language rule base composed of practical control experience; a fuzzy inference part, which enables knowledge-based reasoning and decision-making; and a defuzzification part, which transforms the inferred fuzzy control set into a definite control output.

[0043] The fuzzy controller used in this invention contains four control parameters: two quantization factors and two scaling factors. The quantization factor of the fuzzy controller is K. ex K ecxThe function of the quantization factor is to transform the clear variables from the actual physical domain to the fuzzy domain. The scaling factors are K1 and K2, which transform the fuzzy domain to the physical domain. The quantization factor and the scaling factor realize the function of matching the fuzzy domain and the physical domain, and also have a certain adjustment function. In this embodiment, the fuzzy domains of the input and output variables of the fuzzy controller are both [-6, 6].

[0044] Given a mapping on the fuzzy universe of discourse U:

[0045] A: U→[0,1], x→μ A (x)

[0046] Then set A is a fuzzy set (fuzzy subset) on the fuzzy universe of discourse U; using μ A (x) represents the degree to which each element x in U belongs to set A, and is called the membership function of element x belonging to fuzzy set A. When x is a definite element x0, μ is called... A (x) represents the membership degree of element x0 to fuzzy set A. μ A The closer the value of (x) is to 0, the lower the degree to which x belongs to A, while μ A The closer the value of (x) is to 1, the higher the degree to which x belongs to A. The five common basic membership functions are: triangular, bell-shaped, trapezoidal, sigmoid, and Gaussian. Considering the relatively small range of input variation in the fuzzy controller, the triangular membership function is chosen for the input variables due to its relatively high sensitivity. The output membership functions are all triangular in the middle and Gaussian at both ends, which not only provides sensitivity but also smoother output at the boundaries, exhibiting good stability.

[0047] In the FPEG system, the fuzzy subsets are {NB, NM, NS, ZE, PS, PM, PB}; the Mamdani-type inference algorithm is selected, and its fuzzy implication relations can be obtained through the Cartesian product of fuzzy sets A and B. The control rules are as follows: Figure 4 As shown.

[0048] In summary, the fuzzy PI controller uses the initial error e0 and the error change ec0 as two input variables, and derives ΔK through fuzzification, fuzzy inference, and defuzzification. p ΔK i The obtained ΔK p ΔK i Adding the original PI parameters to the given parameters yields the real-time PI control parameters, thus controlling the controlled object. A fuzzy PI controller based on this principle is shown below. Figure 5 As shown.

[0049] S3. Apply the fuzzy PI controller to the dynamic model to form a double closed-loop feedback to control the piston stroke of the FPEG system;

[0050] In this embodiment, based on the control objective and the FPEG system structure, the entire control system is designed as a dual closed-loop feedback model, such as... Figure 6 As shown, a fuzzy PI controller is applied to the dynamic model, using a fuzzy PI controller in the outer loop (position control loop) and a PI controller in the inner loop (velocity control loop). The input signal to the fuzzy PI controller in the outer loop is the position deviation e between the actual stroke of the piston motion assembly and the target trajectory. x Its output signal is the speed command; the input signal of the inner loop PI controller is the deviation e between the actual speed of the piston motion assembly and the speed command. v Its output signal is a current command.

[0051] Next, control simulation is performed, and the simulation diagram is as follows. Figure 7 As shown. By adjusting the quantization factor and proportional factor of the outer loop fuzzy controller, piston stroke control of the FPEG system can be achieved.

[0052] S4. Optimize the parameters of the fuzzy PI controller using a genetic algorithm.

[0053] In this embodiment, to meet the higher output requirements of the FPEG system, a GA (Genetic Algorithm) is used to optimize the parameters of the fuzzy PI controller. First, the variables to be optimized and the objective function (i.e., the fitness function) are determined, and then GA optimization is performed. First, the variables to be optimized are binary encoded. Then, based on the fitness values ​​of each individual, a selection operation is performed, followed by crossover and mutation operations to generate the next generation. The above evaluation, selection, crossover, and mutation process is repeated until the convergence condition or the required number of iterations is met, such as... Figure 8 As shown.

[0054] The specific optimization parameters (i.e., the variables to be optimized) are: the quantization factor and the scaling factor of the fuzzy PI controller in the outer loop. The integrated time absolute error criterion (ITAE criterion) is used as the fitness function for optimization. This criterion reflects both the magnitude of the error (control accuracy) and the speed of error convergence, thus addressing both the system's rapid response and steady-state performance. Its formula is: GA optimization simulation process as follows Figure 9 As shown, after a certain number of iterations, the optimal solution, i.e., the optimized fuzzy parameters, is obtained, leading to a better control scheme.

[0055] The evaluation metrics corresponding to the simulation results are shown in Table 1 and Figure 10 As shown. The simulation effect is as follows. Figure 11 , 12As shown in Figure 13, for the target sine curve input to the FPEG system, the fuzzy control method, compared with the traditional PI control method, accelerates the response speed, improves steady-state accuracy, reduces output error, and also reduces phase deviation. The fuzzy PI controller with GA-optimized outer loop can achieve piston stroke trajectory control more quickly and accurately.

[0056] Table 1 Evaluation Indicators for Different Control Methods

[0057] index 0.0946 0.00136 0.00123 0.00113

[0058] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. An optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control, characterized in that, include: Construct the dynamic equations of the FPEG system under its operating conditions, and obtain the dynamic model based on the dynamic equations. A fuzzy PI controller is established, comprising a fuzzy controller and a PI controller, wherein the fuzzy controller is used to adjust the proportional parameter of the PI controller. and integral parameters ; A fuzzy PI controller is applied to the dynamic model to form a double closed-loop feedback to control the piston stroke of the FPEG system; The parameters of the fuzzy PI controller are optimized using a genetic algorithm. The piston motion assembly of the FPEG system can be considered as a mass-spring-damped system, and its overall dynamic equation is: By controlling the current To control This controls the FPEG system, ensuring the piston motion follows a sinusoidal trajectory; where, m Mass of moving components, in kg; The piston's trajectory; The pressure of the gas from the left cylinder is in N (N). The pressure of the gas from the right cylinder, in N; Electromagnetic resistance, in N; This is the sum of the frictional forces acting on the piston and the moving part assembly.

2. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 1, characterized in that, When the FPEG system is operating in electric motor mode, the piston reciprocates linearly within the cylinder. The control objective of the electric motor is to make the piston's trajectory track a sinusoidal signal of a certain frequency.

3. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 1, characterized in that, By combining the fuzzy controller and the PI algorithm through the input-output relationship between the fuzzy controller and the PI algorithm, a fuzzy PI controller is formed.

4. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 3, characterized in that, The input signal of the fuzzy controller is the positional deviation of the piston motion assembly stroke. and the rate of change of position deviation The output signal is the parameter adjustment value of the outer loop PI controller. , The quantization factor of the fuzzy controller is , The scaling factor is , .

5. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 3, characterized in that, The fuzzy subset of the input variables of the fuzzy controller is {negative large NB, negative medium NM, negative small NS, zero ZE, positive small PS, positive medium PM, positive large PB}, and the membership functions of the input variables are all triangular, and the membership functions of the output variables are all triangular with Gaussian shape at both ends. The fuzzy controller adopts the Mamdani inference method to set control rules and express the rule table in the form of if-then.

6. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 3, characterized in that, The dynamic model applies a fuzzy PI controller to the outer loop and a PI controller to the inner loop, forming a dual closed-loop feedback control. The fuzzy PI controller outputs speed control commands for the piston motion assembly, while the PI controller outputs current for the piston motion assembly. instruction.

7. The optimized control method for a free-piston internal combustion generator power system based on fuzzy PI control according to claim 1, characterized in that, The GA genetic algorithm is used to optimize the quantization factor and scaling factor of the fuzzy PI controller in the outer loop.

Images