A method and device for identifying parameters of an electromechanical platform based on multi-source disturbance compensation

Abstract

The application relates to the technical field of system parameter identification, and discloses a kind of electromechanical platform parameter identification method and device based on multi-source disturbance compensation, its method includes, establishing nonlinear dynamics equation, small perturbation linearization is carried out at multiple operating points, and the original dynamics linear model of parameter to be determined is obtained;Active introduction of closed-loop controller, realize the robust stabilization of nonlinear system theoretical model at each operating point;Given identification input signal excitation, the system state quantity is measured using two independent sensors, and the corresponding post-processing is carried out on the measurement signal in combination with the sensor characteristics;Based on known input excitation signal and output signal, design expected transfer function, call identification function to identify the system.The application has the effect of improving the parameter identification accuracy of nonlinear system, enhancing the applicability of system parameter identification scheme in nonlinear aspect, strong engineering practicability and migration, can provide effective guidance and inspiration for the accurate identification demand of similar system under disturbance condition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of system parameter identification, in particular to a mechanical and electrical platform parameter identification method and device based on multi-source disturbance compensation. BACKGROUND

[0002] The traditional system identification scheme only needs to determine the positions of the collected input signals and output signals in the system, and then determine the corresponding system model parameters according to the identification results, and is a linear model identification scheme, that is, it must satisfy the assumption that the system is a linear time-invariant system. However, the typical second-order nonlinear system does not satisfy the above assumption, has the disadvantages of high computational complexity, higher requirements for the richness of data collection and signal quality, and is prone to overfitting problems.

[0003] In particular, most unmanned systems are equipped with low-cost MEMS sensors, and the data collected by the sensors have low quality, so when the system has nonlinear characteristics, it cannot be ignored. In addition, considering the actual operating conditions and the constraints of low-cost MEMS sensors, when performing system parameter identification experiments, the collected signals will inevitably be affected by low sampling rate, large measurement noise, and the presence of internal and external disturbances and other adverse complex factors, resulting in reduced identification accuracy. Therefore, the applicability of the traditional system identification scheme is poor and cannot be directly used.

[0004] In view of the above related technologies, the inventors found that the existing system identification scheme has the problems of not being applicable to nonlinear system parameter identification and reduced identification accuracy. SUMMARY

[0005] In order to improve the parameter identification accuracy of a nonlinear system, the present application provides a mechanical and electrical platform parameter identification method and device based on multi-source disturbance compensation.

[0006] In a first aspect, the present application provides a mechanical and electrical platform parameter identification method based on multi-source disturbance compensation.

[0007] The present application is realized by the following technical solutions:

[0008] A mechanical and electrical platform parameter identification method based on multi-source disturbance compensation, comprising the following steps,

[0009] Establishing a nonlinear system theoretical model and performing small perturbation linearization at multiple operating points to determine the to-be-identified parameters after linearization;

[0010] Building an active disturbance rejection controller to realize robust stabilization of the nonlinear system theoretical model at each operating point through parameter tuning;

[0011] Based on the angle sensor and the angular velocity sensor, the corresponding pitch angle signal and the pitch angular rate signal are collected after the given sinusoidal sweep voltage excitation signal is additionally increased under the voltage of the current working point of the motor, until a plurality of groups of preset pitch angles are obtained under the corresponding input voltage signal and the output pitch angle signal and the output pitch angular rate signal;

[0012] The output pitch angle signal and the output pitch angular rate signal are filtered and processed to obtain a target input voltage signal and corresponding target output pitch angle signal and target output pitch angular rate signal;

[0013] The target input voltage signal and the corresponding target output pitch angle signal and the target output pitch angular rate signal are input into the nonlinear system theoretical model, the to-be-identified parameters are combined, an identification function is called to perform parameter identification of the linear system model of the nonlinear system theoretical model at a plurality of working points, and a target parameter is obtained. The target parameter is used for an electromechanical turntable system model.

[0014] In a preferred example, the application can be further configured to:

[0015] A two-degree-of-freedom active disturbance rejection controller including a nominal controller and a disturbance observer is designed.

[0016] The nominal controller is used to achieve the preliminary tracking performance requirements of the closed-loop control system, the disturbance observer is used to online estimate the uncertainty of the internal parameters of the nonlinear system theoretical model and the unknown bounded disturbance in the external environment, and compensate in the total output of the active disturbance rejection controller.

[0017] In a preferred example, the application can be further configured to: the expression of the active disturbance rejection controller u includes,

[0018]

[0019] In the formula, t is time; x is the angle of the electromechanical platform; is the angular velocity of the electromechanical platform; x d is the input expected angle; is the input expected angular velocity; is the input expected angular acceleration; u0(t) is the nominal controller output; is the disturbance observer output; is the initial guess parameter of the nonlinear system theoretical model; J p is the known moment of inertia parameter of the nonlinear system theoretical model; k p , k d and T are parameters to be optimized.

[0020] The application can be further configured in a preferred example that the step of filtering the output pitch angle signal and the output pitch rate signal comprises,

[0021] For multiple sets of output pitch angle signals, a Savitzky-Golay smoothing digital filter is used for upsampling processing.

[0022] For multiple sets of output pitch rate signals, a low-pass filter and linear function fitting are combined for processing.

[0023] The application can be further configured in a preferred example that the expected transfer function comprises a transfer function of a to-be-identified pitch channel dynamics model of a given voltage signal-pitch angle signal and a transfer function of a to-be-identified pitch channel dynamics model of a given voltage signal-pitch rate signal.

[0024] The expression of the transfer function of the to-be-identified pitch channel dynamics model of the given voltage signal-pitch angle signal comprises,

[0025]

[0026] The expression of the transfer function of the to-be-identified pitch channel dynamics model of the given voltage signal-pitch rate signal comprises,

[0027]

[0028] wherein s is a differential operator of the transfer function, J p is a known moment of inertia parameter of the nonlinear system theoretical model, D p ,K sp are initial guess parameters of the nonlinear system theoretical model respectively are accurate estimated values of the initial guess parameters, and K3 is a gain term.

[0029] The application can be further configured in a preferred example that the identification function uses a tfest function of Matlab.

[0030] In a second aspect, the application provides an electromechanical platform parameter identification device based on multi-source disturbance compensation.

[0031] The application is achieved by the following technical solutions:

[0032] An electromechanical platform parameter identification device based on multi-source disturbance compensation comprises,

[0033] A nonlinear model module is configured to establish a nonlinear system theoretical model, perform small disturbance linearization at multiple working points, and determine to-be-identified parameters after linearization.

[0034] A robust module is configured to build an active disturbance rejection controller, and to realize robust stabilization of the nonlinear system theoretical model at each of the operating points through parameter tuning.

[0035] A collection module is configured to collect, based on an angle sensor and an angular velocity sensor, a corresponding pitch angle signal and a pitch angular velocity signal of the motor after an additional given sinusoidal sweep voltage excitation signal is added to a voltage at a current operating point, until a plurality of sets of output pitch angle signals and output pitch angular velocity signals under a plurality of sets of preset pitch angles under corresponding input voltage signals are obtained.

[0036] A filtering module is configured to perform filtering processing on the output pitch angle signal and the output pitch angular velocity signal, and to integrate a target input voltage signal and corresponding target output pitch angle signal and target output pitch angular velocity signal.

[0037] An identification module is configured to design a desired transfer function, to input the target input voltage signal and the corresponding target output pitch angle signal and the target output pitch angular velocity signal into the nonlinear system theoretical model, to combine the to-be-identified parameters, to call an identification function to perform parameter identification of a linear system model of the nonlinear system theoretical model at a plurality of operating points, and to obtain target parameters, the target parameters being used for an electromechanical turntable system model.

[0038] In a third aspect, the present application provides a computer device.

[0039] The present application is achieved by the following technical solutions:

[0040] The computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of any one of the above-mentioned electromechanical platform parameter identification methods based on multi-source disturbance compensation when executing the computer program.

[0041] In a fourth aspect, the present application provides a computer readable storage medium.

[0042] The present application is achieved by the following technical solutions:

[0043] The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of any one of the above-mentioned electromechanical platform parameter identification methods based on multi-source disturbance compensation.

[0044] In a fifth aspect, the present application provides a computer program product.

[0045] The present application is achieved by the following technical solutions:

[0046] A computer program product comprises a computer program which, when executed by a processor, implements the steps of any of the above-mentioned methods for parameter identification of an electromechanical platform based on compensation of multi-source disturbances.

[0047] In summary, compared with the prior art, the technical scheme provided by the application has at least the following beneficial effects:

[0048] The nonlinear system theoretical model is established, and small perturbation linearization is performed at multiple operating points to obtain linearized models of a single nonlinear system model at different operating points, and the to-be-identified parameters after linearization are determined; an active anti-disturbance controller is built, an active closed-loop controller is introduced, and parameter tuning is performed to realize robust stabilization of the nonlinear system theoretical model at each operating point, which is beneficial to realizing robust stabilization at a specified operating point under the condition that the initial parameters of the model are inaccurate; based on an angle sensor and an angular velocity sensor, corresponding pitch angle signals and pitch angular rate signals of the motor at the current operating point after an additional given sinusoidal sweep voltage excitation signal is added to the voltage are collected until a plurality of sets of output pitch angle signals and output pitch angular rate signals under a plurality of sets of preset input voltages are obtained; the frequency domain information of the sinusoidal sweep voltage excitation signal is more sufficient, the pitch angle and the pitch angular rate are zero-order and first-order state quantities of the system, and can fully and sufficiently represent the current motion mode of the system, thereby providing sufficient and effective model response information for identification, and being beneficial to improving the subsequent identification accuracy; the output pitch angle signals and the output pitch angular rate signals are filtered to reduce the influence of large noise, existing internal and external disturbances and other adverse complex factors on the collected signals, and are beneficial to improving the subsequent identification accuracy; an expected transfer function is designed, the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angular rate signal are input into the nonlinear system theoretical model, the to-be-identified parameters are combined, an identification function is called to perform parameter identification of the linear system model of the nonlinear system theoretical model at multiple operating points, and target parameters are obtained, the target parameters are used for the electromechanical turntable system model, and the nonlinear system can be robustly and accurately stabilized at the corresponding operating point; the parameter identification accuracy of the nonlinear system is improved, the applicability of the system parameter identification scheme in the nonlinear aspect is enhanced, the purpose of high-precision parameter identification of a widely existing disturbed nonlinear second-order electromechanical platform is achieved; the identification result of the nonlinear system identification scheme of the present scheme is the real physical parameters of the model, has strong interpretability, can directly assist researchers in establishing an accurate system model, considers the identification requirements in the engineering application background, does not require a large amount of high-quality input-output data, effectively processes the multi-source disturbance factors existing in the actual environment, and has strong engineering landing property. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1A main flowchart of a multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0050] Figure 2 A system nonlinear dynamics modeling diagram of an AERO turntable according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to another example embodiment of the present application.

[0051] Figure 3 A result diagram of a stabilization experiment of an AERO turntable at four different working points according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to another example embodiment of the present application.

[0052] Figure 4 A comparison diagram of original signals and processed signals obtained from an identification experiment according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0053] Figure 5 A result comparison diagram of model responses and actual responses at four different working points according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0054] Figure 6 A statistical result distribution diagram of experimental data at four different working points according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0055] Figure 7 A comparison diagram of a theoretical curve and a fitted experimental curve of an angle-gravity moment coefficient of a nonlinear model according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0056] Figure 8 A result comparison diagram of model responses and actual responses of two different sensors according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0057] Figure 9 A result comparison diagram of model responses and actual responses of two similar turntables of the same type at the same working point according to the multi-source disturbance compensation based parameter identification method of an electromechanical platform according to an example embodiment of the present application.

[0058] Figure 10 A structure block diagram of a multi-source disturbance compensation based parameter identification device according to an example embodiment of the present application. DETAILED DESCRIPTION

[0059] The specific embodiments are merely explanatory of this application, and are not intended to limit this application, and those skilled in the art can make modifications to the embodiments without creative contribution after reading the specification, as long as the modifications are within the scope of the claims of this application.

[0060] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative contribution are within the scope of protection of the present application.

[0061] In addition, the term "and / or" in this paper is only to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in this paper generally represents an "or" relationship between the associated objects unless otherwise specified.

[0062] The embodiments of the present application will be described in further detail below with reference to the drawings of the specification.

[0063] Referring to Figure 1 The embodiments of the present application provide a kind of based on multi-source disturbance compensation electromechanical platform parameter identification method, the main steps of the described method are described as follows.

[0064] S1: establish a nonlinear system theoretical model, and linearize at multiple operating points Small disturbance, determine the to-be-identified parameters after linearization;

[0065] S2: build an active disturbance rejection controller, and realize the robust stabilization of the nonlinear system theoretical model at each operating point by parameter tuning;

[0066] S3: based on angle sensor and angular velocity sensor, respectively collect the corresponding pitch angle signal and pitch angular rate signal after the additional given sinusoidal sweep voltage excitation signal is added to the voltage of the current operating point of motor, until a plurality of preset pitch angles are obtained Output pitch angle signal and output pitch angular rate signal under corresponding input voltage signal;

[0067] S4: filter processing is carried out on the output pitch angle signal and the output pitch angular rate signal, and the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angular rate signal are integrated to obtain;

[0068] S5: design a desired transfer function, input the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angle rate signal into the nonlinear system theory model, combine the to-be-identified parameters, call an identification function to perform parameter identification of a linear system model of the nonlinear system theory model at multiple working points, and obtain target parameters, the target parameters being used for an electromechanical turntable system model.

[0069] In an embodiment, the step of building the active disturbance rejection controller comprises,

[0070] designing a two-degree-of-freedom active disturbance rejection controller comprising a nominal controller and a disturbance observer;

[0071] The nominal controller is used to achieve preliminary tracking performance requirements of a closed-loop control system, and the disturbance observer is used to perform online estimation on the uncertainty of internal parameters of the nonlinear system theory model and unknown bounded disturbances in the external environment and perform compensation in the total output of the active disturbance rejection controller.

[0072] In an embodiment, the expression of the active disturbance rejection controller u comprises,

[0073]

[0074] In the formula, t is time; x is an angle of an electromechanical platform; is an angular velocity of the electromechanical platform; x d is an input desired angle; is an input desired angular velocity; is an input desired angular acceleration; u0(t) is a nominal controller output; is a disturbance observer output; is an initial guess parameter of the nonlinear system theory model; J p is a known moment of inertia parameter of the nonlinear system theory model; k p , k d and T are parameters to be optimized.

[0075] In an embodiment, the step of filtering the output pitch angle signal and the output pitch angle rate signal comprises,

[0076] For multiple sets of output pitch angle signals, a Savitzky-Golay smoothing digital filter is used for upsampling processing;

[0077] For multiple sets of output pitch angle rate signals, a combination of low-pass filtering and linear function fitting for detrending is used for processing.

[0078] In an embodiment, the desired transfer function comprises a transfer function of a pitch channel dynamics model to be identified for a given voltage signal-pitch angle signal and a transfer function of a pitch channel dynamics model to be identified for a given voltage signal-pitch angle rate signal.

[0079] The expression of the transfer function of the pitch channel dynamics model to be identified for a given voltage signal-pitch angle signal comprises,

[0080]

[0081] The expression of the transfer function of the pitch channel dynamics model to be identified for a given voltage signal-pitch angle rate signal comprises,

[0082]

[0083] where s is a differential operator of the transfer function, J p is a known moment of inertia parameter of the nonlinear system theoretical model, D p ,K sp are initial guess parameters of the nonlinear system theoretical model respectively are accurate estimated values of the initial guess parameters, and K3 is a gain term.

[0084] In an embodiment, the identification function employs a Matlab tfest function.

[0085] In an embodiment, a multi-source disturbance compensation based parameter identification method for an electromechanical platform further comprises,

[0086] The target parameter is evaluated from the perspective of statistical analysis, theoretical fitting, and data comparison.

[0087] The specific descriptions of the above embodiments are as follows.

[0088] Referring to Figure 2 , a nonlinear dynamics equation is established based on an AERO turntable to obtain a nonlinear system theoretical model:

[0089]

[0090] Small perturbation linearization is performed at a plurality of given working points (θ0, τ p0 ), and the nonlinear model is converted into a plurality of linearized models:

[0091] K sp = mglcosθ0, Δθ = θ - θ0, Δτ p = τ p - τ p0

[0092] The plurality of linear system models actually correspond to a certain nonlinear system near a plurality of different operating points, i.e., linearization models of a single nonlinear system model at different operating points. The plurality of linearization models are second-order linear dynamic models.

[0093] A small-perturbation linear dynamic model containing parameters J p ,D p ,K sp is finally obtained, J p has a clear physical meaning and can be obtained by actual measurement, i.e., the undetermined parameters of the transfer function model are D p and K sp , and the to-be-identified parameters after linearization are determined.

[0094] Then, based on the actuator model, an active disturbance rejection controller is built.

[0095] The system model after linearization near the operating point θ0is:

[0096]

[0097] Define x = Δθ and u = Δτ p , and the typical second-order system can be written as:

[0098]

[0099] Based on the feedback linearization idea, the active disturbance rejection controller can be defined as:

[0100]

[0101] wherein, is the initial guessed parameter of the model.

[0102] Further, the active disturbance rejection controller includes a nominal controller and a disturbance observer. The active disturbance rejection controller is two-degree-of-freedom, wherein the nominal controller is used to realize the preliminary tracking performance requirements of the closed-loop control system, the disturbance observer is used to online estimate the uncertainty of the internal parameters of the theoretical model of the nonlinear system and the unknown bounded disturbance in the external environment, and compensate in the total output of the active disturbance rejection controller, so as to finally realize the robust stabilization of the model at the specified operating point under the condition that the initial parameters of the model are inaccurate.

[0103] The nominal controller u0is designed as:

[0104]

[0105] The nominal controller u0is a typical PD controller with acceleration feedforward signal.

[0106] The disturbance observer is a low-order uncertainty and disturbance estimator, designed as:

[0107]

[0108] By good tuning of the parameters k p ,k d ,T of the above controller, the parameters k p ,k d of the PD controller are first determined to achieve system stability, at which time the transient performance is good and the steady-state error is large, and then the value of T of the disturbance observer is adjusted to reduce the value of T to prevent the transient overshoot from being too large, so as to achieve a balance between transient performance and steady-state performance, that is, the robust stabilization of the theoretical model of the nonlinear system at each working point. According to the allowed working interval information of the actual electromechanical turntable system, a plurality of working points are uniformly designed within the interval, the turntable pitch angle is stably and accurately controlled to a certain constant value, which is taken as a working point, and the pitch angle is sequentially given as 0.1 rad, 0.2 rad, 0.3 rad and 0.4 rad, which are taken as the working points of the system.

[0109] The original nonlinear dynamic equation is linearized by small perturbation to obtain a second-order linear dynamic model at each working point. The stabilization results at different pitch angles are shown in Figure 3 .

[0110] Further, a dynamic model identification experiment is performed, in which an angle sensor and an angular velocity sensor are used for measurement, respectively, the corresponding pitch angle signal and pitch angular rate signal are collected after an additional given sinusoidal sweep voltage excitation signal is added to the voltage at the current working point of the motor, the system state quantity information is obtained, and the identification signal is obtained after corresponding signal post-processing.

[0111] Specifically, the dynamic model identification experiment includes sequentially giving the pitch angle as 0.1 rad, 0.2 rad, 0.3 rad and 0.4 rad, and performing corresponding experiments near the four different pitch angle working points. The AERO turntable is started, and the active disturbance rejection controller is first called to stabilize the pitch angle near the working point, and the corresponding voltage value V p0 after stabilization is recorded. Then, the input voltage signal of the AERO turntable is switched to the following form:

[0112]

[0113] In the formula, ΔV p is the amplitude of the sinusoidal sweep voltage excitation signal. Considering the condition of small perturbation linearization, ΔV p should not be too large, and in order to distinguish from the disturbance, ΔV pIt also cannot be very small. [f0, f1] is the frequency interval of the sinusoidal sweep voltage excitation signal, for [f0, f1], the frequency domain interval of interest is selected or the interval covering the actuator execution bandwidth, and T is the duration of the sinusoidal sweep signal.

[0114] After determining the sinusoidal sweep voltage excitation signal, when the turntable is stable near the working point, the motor additionally adds the sinusoidal sweep voltage excitation signal, and based on the angle encoder, the input voltage signal and the output pitch angle signal after the voltage signal switching are measured and recorded; based on the MEMS gyroscope, the input voltage signal and the output pitch angle rate signal after the voltage signal switching are measured and recorded.

[0115] Through different sensors, corresponding pitch angle signals and pitch angle rates are collected, each group is repeated multiple times, and multiple groups of input voltage signals and output pitch angle signals and output pitch angle rate signals after voltage signal switching are obtained, which is beneficial to improve the subsequent parameter identification accuracy.

[0116] The angle sensor and the angular velocity sensor have low cost and can be widely applied to the motion state measurement of electromechanical systems; from the theoretical dynamic model, the (angular) displacement and the (angular) velocity are the zero-order and first-order state quantities of the system, which can fully characterize the current motion mode of the system, thereby providing sufficient and effective model response information for identification.

[0117] Reference Figure 4 Since the photoelectric angle encoder carried by the AERO turntable has a low sampling rate, in order to ensure that the identification result has high accuracy, the output pitch angle signal and the output pitch angle rate signal collected are filtered, the original measurement signal is up-sampled, and the target input voltage signal and the corresponding target output pitch angle signal and the target output pitch angle rate signal are integrated.

[0118] Considering that the noise contained in the original encoder data is relatively small, the low sampling rate signal of the angle encoder can be directly smoothed by the Savitzky-Golay filter, which can basically not lose the information of the original signal. The low-cost MEMS gyroscope carried by the AREO turntable has large noise, and due to the characteristics of the gyroscope sensor, it has a constant drift, so the high-frequency noise and constant drift sampling signal of the MEMS gyroscope is removed by low-pass filtering to remove high-frequency noise, and combined with linear function fitting to remove the constant drift of the gyroscope, to obtain smooth and constant drift-free data.

[0119] The processed signals are compared as Figure 4As shown in the figure, the abscissa represents time (s) and the ordinate represents the pitch angle (rad) or the pitch angle rate (rad / s). By giving reasonable filter parameters, the up-sampling of the low sampling rate data of the encoder and the filtering processing of the low-cost MEMS gyroscope can be effectively realized without substantially losing the amplitude and phase information.

[0120] After the input voltage signal and the processed output pitch angle (rate) signal are known, the results are taken as input-output data, and the model parameter identification of the system can be performed. By designing the desired transfer function, the target input voltage signal and the corresponding target output pitch angle signal and the target output pitch angle rate signal are input into the nonlinear system theoretical model, the identification function is called to perform the parameter identification of the linear system model of the nonlinear system theoretical model at multiple operating points in combination with the to-be-identified parameters, and the target parameters for the electromechanical turntable system model are obtained.

[0121] According to the linearized dynamic model, a transfer function model of the motor voltage to the pitch angle and the pitch angle rate is established.

[0122] The transfer function structure of the to-be-identified pitch channel dynamic model of the given voltage signal to the pitch angle is:

[0123]

[0124] For the to-be-identified pitch channel dynamic model of the given voltage signal to the pitch angle rate, it only needs to be differentiated from the above model, and the transfer function structure thereof is:

[0125]

[0126] wherein,

[0127] The data measured by different sensors are subjected to transfer function modeling to obtain the desired transfer function. The parameter identification of the pitch channel dynamic model is performed by subtracting the initial value of the operating point from the voltage signal and the processed output pitch angle (rate) signal, respectively, to obtain the input data-output data of the to-be-identified model, and then the identification function tfest is called to perform the model parameter identification of the system by giving the to-be-identified transfer function structure in the idtf function of the System Identification Toolbox of Matlab, so as to obtain the optimal fitting parameter identification result, which can ensure that the nonlinear system is robustly and accurately stabilized at the corresponding operating point.

[0128] In the real object identification experiment of the pitch channel dynamic model, the comparison between the identification result fitting curves and the actual responses measured by the two sensors is as shown in the figure. Figure 5 .

[0129] Finally, the identified results: the transfer function identification parameters K1 and K2 at four different pitch angle working points are comprehensively evaluated from the aspects of statistical analysis, theoretical fitting, data comparison, etc. Considering the inertia parameters J p of the AERO turntable are easy to measure and change little, they can be regarded as known quantities, and then the friction coefficient D p and the gravity torque coefficient K sp of the turntable can be calculated as follows:

[0130] D p = K1J p

[0131] K sp = K2J p

[0132] From the perspective of statistical analysis, after multiple repeated experiments on the same AERO device, the average value and standard deviation of the identification parameters of each group are calculated by statistical methods.

[0133] The average value and standard deviation of the friction coefficient at each working point in the real object identification results are shown in Table 1, and the average value and standard deviation of the gravity torque coefficient at each working point are shown in Table 2. Figure 6 Figure 7

[0134] From the perspective of multiple independent sensors, different sensor methods are used to measure the AERO device to avoid the contingency caused by a single sensor. The average value and standard deviation of the identification parameters of the two sensors at each working point are shown in Table 3 and Table 4, respectively. It can be found that the results obtained by the two sensors are similar, which verifies the reliability of the experiment from the perspective of the sensor and reduces the measurement error of a single sensor. Figure 6 Figure 7

[0135] The dispersion coefficient is calculated by the following formula for comparison of multiple independent experiments:

[0136]

[0137] The comparison of the dispersion coefficients can show the representativeness and stability of the average indicators of different populations. The smaller the dispersion coefficient, the better the representativeness of the average indicator.

[0138] Referring to Figure 8 ​​​​The dispersion coefficients of the friction coefficient and the gravity moment coefficient of the two sensors at each working point, and the standard deviation of the identified parameters are of the order of 10-4, and the dispersion coefficients are less than 0.1, regardless of the experimental results measured by either sensor, indicating that the experimental results of each group are reliable and excluding accidental factors of the experiment.

[0139] From a theoretical point of view, the gravity moment coefficient and the pitch angle are approximately in a cosine relationship, while the friction coefficient and the pitch angle are approximately constant. The fitting relationship and whether it is approximately constant can be determined by comparing the cosine fitting of the relationship between each gravity moment coefficient and the pitch angle, and comparing the friction coefficient at each working point. The reliability of the model parameters is theoretically verified.

[0140] Referring to Figure 9 From the equipment point of view, repeated experiments on the same type of AERO equipment produced in the same batch are compared to exclude the randomness of a single device. The comparison results of the identified model response and the actual response of two similar turntables of the same type at the same working point are shown in Figure 9 As can be seen from the results, the parameters of the two AERO devices are basically similar but not exactly the same, and the identified results can be verified to be reasonable and reliable.

[0141] In summary, a multi-source disturbance compensation-based electromechanical platform parameter identification method establishes a nonlinear system theoretical model, linearizes at multiple operating points to obtain linearized models of a single nonlinear system model at different operating points, and determines the parameters to be identified after linearization; an active disturbance rejection controller is built to actively introduce a closed-loop controller, and parameter tuning is used to achieve robust stabilization of the nonlinear system theoretical model at each operating point, which is beneficial for achieving robust stabilization at a specified operating point under the condition that the initial parameters of the model are inaccurate; based on angle sensors and angular velocity sensors, the corresponding pitch angle signal and pitch angular rate signal are collected after the motor at the current operating point is additionally increased by a given sinusoidal sweep voltage excitation signal under the voltage, until a plurality of preset pitch angles are obtained under the corresponding input voltage signal and the output pitch angle signal and the output pitch angular rate signal, the frequency domain information of the sinusoidal sweep voltage excitation signal is more sufficient, and the pitch angle and the pitch angular rate are the zero-order and first-order state quantities of the system, which can fully and sufficiently represent the current motion mode of the system, thereby providing sufficient and effective model response information for identification, which is beneficial for improving the subsequent identification accuracy; the output pitch angle signal and the output pitch angular rate signal are filtered to reduce the influence of noise, internal and external disturbances and other adverse complex factors on the collected signals, which is beneficial for improving the subsequent identification accuracy; an expected transfer function is designed, the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angular rate signal are input into the nonlinear system theoretical model, the to-be-identified parameters are combined, an identification function is called to identify the parameters of the linear system model of the nonlinear system theoretical model at multiple operating points, and target parameters are obtained, which are used for electromechanical turntable system model, which can ensure that the nonlinear system is robust and accurately stabilized at the corresponding operating point; thereby improving the parameter identification accuracy of the nonlinear system, enhancing the applicability of the system parameter identification scheme in the nonlinear aspect, and achieving the purpose of high-precision parameter identification of a widely existing disturbed nonlinear second-order electromechanical platform; the identification result of the nonlinear system identification scheme of the present scheme is the real physical parameters of the model, which has strong interpretability and can directly assist researchers in establishing an accurate system model, while considering the identification requirements in the engineering application background, without a large amount of high-quality input-output data, effectively handling the multi-source disturbance factors existing in the actual environment, and having strong engineering landing property.

[0142] The method for identifying parameters of an electromechanical platform based on multi-source disturbance compensation is suitable for a nonlinear system model under conditions of poor state measurement signal quality and motion disturbed by uncertain disturbances, and by actively introducing a closed-loop robust controller, combining a small disturbance assumption, and approximating the original nonlinear model by multiple linear models at different working points and parameters to be identified, the effect of accurately identifying the nonlinear system by using a linear identification scheme is achieved, and the method is suitable for a nonlinear system model under bad conditions such as large measurement noise and strong model nonlinearity, and has the advantage of low calculation complexity.

[0143] The active anti-disturbance controller based on a low-order disturbance estimator and a nominal controller combines initial guessed parameter information of the model, realizes robust stabilization of the turntable at the expected working point to be identified under the condition of large model parameter uncertainty, gives appropriate identification input signal excitation, measures system state quantities (or derivatives of state quantities) by using two independent sensors, and combines sensor characteristics to perform corresponding post-processing on the measurement signals, and combines known input excitation signals to serve as input-output data for system identification.

[0144] The method for identifying parameters of an electromechanical platform based on multi-source disturbance compensation is suitable for a typical disturbed second-order nonlinear dynamic system in actual engineering, and gives an accurate identification scheme closely combining theory and engineering implementation, including four links of modeling, stabilization, identification, and evaluation, and has strong engineering practicability and migration, high identification result accuracy, and can provide effective guidance and inspiration for accurate identification requirements of similar systems under disturbed conditions.

[0145] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0146] Reference Figure 10 The embodiments of the present application also provide a device for identifying parameters of an electromechanical platform based on multi-source disturbance compensation, which corresponds to the method for identifying parameters of an electromechanical platform based on multi-source disturbance compensation in the above embodiment. The device for identifying parameters of an electromechanical platform based on multi-source disturbance compensation comprises,

[0147] A nonlinear model module is configured to establish a nonlinear system theoretical model, linearize the model at multiple working points, and determine parameters to be identified after linearization.

[0148] A robust module is configured to build an active anti-disturbance controller, and realize robust stabilization of the nonlinear system theoretical model at each working point by parameter tuning.

[0149] The collection module is configured to collect, based on the angle sensor and the angular velocity sensor, a corresponding pitch angle signal and a pitch angular velocity signal after an additional given sinusoidal sweep voltage excitation signal is added to a voltage of a current working point of the motor, until a plurality of sets of output pitch angle signals and output pitch angular velocity signals under a plurality of sets of preset pitch angles and corresponding input voltage signals are obtained.

[0150] The filtering module is configured to perform filtering processing on the output pitch angle signal and the output pitch angular velocity signal, and integrate to obtain a target input voltage signal and corresponding target output pitch angle signal and target output pitch angular velocity signal.

[0151] The identification module is configured to design an expected transfer function, input the target input voltage signal and the corresponding target output pitch angle signal and the target output pitch angular velocity signal into the nonlinear system theoretical model, combine the to-be-identified parameters, call an identification function to perform parameter identification of a linear system model of the nonlinear system theoretical model at a plurality of working points, and obtain a target parameter, the target parameter being used for an electromechanical turntable system model.

[0152] The electromechanical platform parameter identification device based on multi-source disturbance compensation further includes,

[0153] The evaluation module is configured to evaluate the target parameter from the perspective of statistical analysis, theoretical fitting, and data comparison.

[0154] Specific limitations of the electromechanical platform parameter identification device based on multi-source disturbance compensation can be referred to the limitations of the electromechanical platform parameter identification method based on multi-source disturbance compensation, which will not be repeated here.

[0155] Each module in the electromechanical platform parameter identification device based on multi-source disturbance compensation can be realized by software, hardware, and a combination thereof, in whole or in part. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform the operations corresponding to each module.

[0156] In an embodiment, a computer device is provided, which can be a server. The computer device comprises a processor, a memory, a network interface and a database connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is used to communicate with the external terminal through the network connection. The computer program is executed by the processor to implement any one of the above-mentioned multi-source disturbance compensation-based parameter identification methods of an electromechanical platform.

[0157] In an embodiment, a computer readable storage medium is provided, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement any one of the above-mentioned multi-source disturbance compensation-based parameter identification methods of an electromechanical platform.

[0158] In an embodiment, a computer program product is provided, comprising a computer program, which is executed by a processor to implement any one of the above-mentioned multi-source disturbance compensation-based parameter identification methods of an electromechanical platform.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware. The computer program can be stored in a non-volatile computer readable storage medium, and includes instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The computer program can include the processes of the above-mentioned embodiments of the methods when executed. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0160] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the functions described above.

Claims

1. A method for parameter identification of an electromechanical platform based on multi-source disturbance compensation, characterized in that, The method comprises the following steps, a nonlinear system theory model is established, and small perturbation linearization is performed at multiple operating points to determine the linearized parameters to be identified; The active disturbance rejection controller is built, and the robust stabilization of the nonlinear system theory model at each working point is realized through parameter tuning, wherein the active disturbance rejection controller The expression of the active disturbance rejection controller is: ； ； ； wherein is time; is the electromechanical platform angle; is the electromechanical platform angular velocity; is the input desired angle; is the input desired angular velocity; is the input desired angular acceleration; is the nominal controller output; is the disturbance observer output; , is the initial guess parameter of the nonlinear system theory model; is the known moment of inertia parameter of the nonlinear system theory model; , and are the parameters to be tuned; Based on the angle sensor and the angular velocity sensor, the corresponding pitch angle signal and the pitch angle rate signal of the motor at the current operating point are collected after the given sinusoidal sweep voltage excitation signal is additionally added to the voltage, until a plurality of sets of output pitch angle signals and output pitch angle rate signals under the corresponding input voltage signals of the preset pitch angle are obtained; The output pitch angle signal and the output pitch angle rate signal are filtered and processed to obtain the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angle rate signal; An expected transfer function is designed, the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angle rate signal are input into the nonlinear system theory model, the parameters to be identified are combined, an identification function is called to perform parameter identification of the linear system model of the nonlinear system theory model at multiple operating points, and target parameters are obtained, which are used for the electromechanical turntable system model.

2. The multi-source disturbance compensation based parameter identification method of an electromechanical platform according to claim 1, wherein, The step of building an active disturbance rejection controller comprises, a two-degree-of-freedom active disturbance rejection controller including a nominal controller and a disturbance observer is designed; The nominal controller is used to achieve the preliminary tracking performance requirements of the closed-loop control system, and the disturbance observer is used to online estimate the uncertainty of the internal parameters of the nonlinear system theory model and the unknown bounded disturbance in the external environment, and compensate in the total output of the active disturbance rejection controller.

3. The multi-source disturbance compensation based parameter identification method of an electromechanical platform according to claim 1, wherein, The step of filtering and processing the output pitch angle signal and the output pitch angle rate signal comprises, For a plurality of sets of output pitch angle signals, Savitzky-Golay smoothing digital filter is used for upsampling processing; For a plurality of sets of output pitch angle rate signals, low-pass filtering and linear function fitting are combined for processing.

4. The multi-source disturbance compensation based parameter identification method of an electromechanical platform according to any one of claims 1-3, characterized in that, The expected transfer function comprises a transfer function of a to-be-identified pitch channel dynamics model of a given voltage signal-pitch angle signal and a transfer function of a to-be-identified pitch channel dynamics model of a given voltage signal-pitch angle rate signal; The expression of the transfer function of the to-be-identified pitch channel dynamics model of the given voltage signal-pitch angle signal comprises, ； The expression of the transfer function of the to-be-identified pitch channel dynamics model of the given voltage signal-pitch angle rate signal comprises, ； wherein is a differential operator of the transfer function, is a known moment of inertia parameter of the nonlinear system theoretical model, are initial guess parameters of the nonlinear system theoretical model, respectively are accurate estimates of the initial guess parameters, is a gain term.

5. The multi-source disturbance compensation based parameter identification method of an electromechanical platform according to claim 4, wherein, The identification function uses the tfest function of Matlab.

6. A multi-source disturbance compensation based electromechanical platform parameter identification device, characterized in that, The steps of the method for performing the parameter identification of the electromechanical platform based on multiple source disturbance compensation according to any one of claims 1 to 5 comprise, a nonlinear model module for establishing a nonlinear system theory model and performing small perturbation linearization at multiple operating points to determine the linearized parameters to be identified; a robust module for building an active disturbance rejection controller to realize robust stabilization of the nonlinear system theory model at each operating point through parameter tuning. The collection module is configured to collect, based on the angle sensor and the angular velocity sensor, a corresponding pitch angle signal and a pitch angular rate signal after an additional given sinusoidal sweep voltage excitation signal is added to a voltage of a current working point of the motor, until a plurality of sets of output pitch angle signals and output pitch angular rate signals corresponding to a plurality of sets of preset pitch angles under input voltage signals are obtained. The filtering module is configured to perform filtering processing on the output pitch angle signal and the output pitch angular rate signal, and integrate to obtain a target input voltage signal and corresponding target output pitch angle signal and target output pitch angular rate signal. The identification module is configured to design an expected transfer function, input the target input voltage signal and the corresponding target output pitch angle signal and target output pitch angular rate signal into the nonlinear system theoretical model, combine the to-be-identified parameters, call an identification function to perform parameter identification of a linear system model of the nonlinear system theoretical model at a plurality of working points, and obtain target parameters, the target parameters being used for a model of the electromechanical turntable system.

7. A computer device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 5.

9. A computer program product, characterised in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 5.

Images