Humanoid robot doubly salient memory motor optimization method

Abstract

The invention discloses a humanoid robot doubly salient memory motor optimization method, relates to the technical field of electric transmission control equipment, and aims to solve the problems that high-coercive-force permanent magnets, low-coercive-force permanent magnets and windings of a humanoid robot doubly salient memory motor are all arranged on a stator side, and the two permanent magnets interfere with each other, so that the space in a stator slot is limited, and the stability of the humanoid robot doubly salient memory motor is influenced. The stator slot area needs to be optimized. The method comprises the following steps: starting from meeting the requirements of magnetic load and torque density performance indexes of the doubly salient memory motor, giving the magnetic load and torque density indexes, and proposing a Pareto weighted objective function to quantify a sample database in multi-attribute decisions of the two indexes so as to obtain an optimal solution; a linear relation between a doubly salient memory motor structure parameter and an optimization performance index is measured through a Pearson coefficient, and sensitivity is introduced to evaluate the comprehensive influence of two permanent magnets on a magnetic load and a torque density optimization target. And finally, the sine performance effect of the no-load back electromotive force after optimization of the doubly salient memory motor is better than that of the no-load back electromotive force before optimization.

Description

Technical Field

[0001] The invention relates to the technical field of electric drive control equipment, in particular to an optimization method for a double-salient-pole memory motor of a humanoid robot. Background Art

[0002] In the motor drive transmission operation of humanoid robot joints, the double-salient-pole memory motor is applied because of its good magnetic field regulation ability. However, the double-salient-pole memory motor has two kinds of permanent magnets, high coercivity permanent magnet and low coercivity permanent magnet, which reduces the risk of permanent magnet demagnetization and overcomes the difficulty of magnetic flux adjustment. It also makes the stator space structure characteristics very complex, which leads to mutual interference between permanent magnets in the stator slots, weakens the magnetic load and torque density performance, and causes poor sinusoidal property of no-load back EMF. Compared with traditional permanent magnet motors, the double-salient-pole memory motor has the important feature of adjustable magnetic flux. It is the core component for realizing electromechanical energy conversion of humanoid robot joint motors. It is in an important position of energy conversion, and plays the role of matching speed and transmitting torque between the prime mover and the working machine or actuator. It is of great significance to ensure the reliable operation of humanoid robots. The control system of the double-salient-pole memory motor has multi-objective, nonlinear and strong coupling problems. The difficulty lies in accurately and quickly establishing mathematical models between multiple optimization objectives and the main structure.

[0003] After searching relevant patents and literature at home and abroad, the optimization design methods of memory motors mainly include genetic algorithms, extended Kalman filter methods, cuckoo algorithms, gray wolf optimization algorithm search, etc. Although there are many motor systems used for the optimization and improvement of the performance of humanoid robot joint motors, no patents and literature have been found for the optimization design of double-salient memory motors for humanoid robot joints. Literature on optimization design methods for similar motors, such as "Multi-objective optimization design of axial magnetic field permanent magnet memory motor based on response surface method", initially selected three influencing factors of soft magnetic ratio, rotor pole number, and air gap length as design factors for orthogonal experiments, and then used the response surface optimization method to select the selected induced electromotive force, total harmonic distortion rate of induced electromotive force, and cogging torque as optimization targets. However, the original induced electromotive force cannot be guaranteed when the cost of permanent magnet materials is reduced. And "Design optimization and characteristic analysis of new double-layer series hybrid permanent magnet memory motor" proposes a set of multi-objective design optimization processes to determine the initial size of the motor for problems such as limited magnetic adjustment range and high magnetic adjustment current amplitude. The invention patent with application number 2019100667535 and name "A method for predicting and optimizing the online magnetic performance of a flux-controlled memory motor based on a nonlinear equivalent variable magnetic network model" obtains the optimal parameter combination from the parameter group at the forefront of the optimal solution set and determines the motor structure, but fails to change the excitation current and narrow the range of the motor structure solution set.

[0004] In summary, the existing optimization design methods can be used in a variety of memory motors, but due to the limitations of estimation speed and accuracy, they have not yet been applied to the double-salient-pole memory motor system of humanoid robot joints. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for optimizing the double-pole memory motor of a humanoid robot. The present invention can ensure the accuracy of the control system, so that it can have good sinusoidal no-load back electromotive force and better speed regulation performance at different electrical angles, and the torque is stably increased.

[0006] The present invention adopts the following technical solutions to solve the above technical problems:

[0007] According to a method for optimizing a double-salient-pole memory motor for a humanoid robot proposed in the present invention, given two indicators, magnetic load and torque density of the double-salient-pole memory motor, a Pareto-weighted objective function is used to quantify a sample database in multi-attribute decision making of the two indicators, magnetic load and torque density of the double-salient-pole memory motor, so as to obtain the optimal solutions of the magnetic load and torque density indicators;

[0008] The Pearson coefficient is used to measure the linear relationship between the structural parameters of the double-pole memory motor and the optimal solution of the magnetic load and torque density indicators. This linear relationship is used to optimize the motor's magnetic flux intensity and real-time adjustment of the torque as the load changes.

[0009] As a further optimization scheme of the humanoid robot double-pole memory motor optimization method described in the present invention, it also includes: introducing sensitivity to evaluate the comprehensive influence of high-coercive force permanent magnets and low-coercive force permanent magnets on the magnetic load and torque density optimization targets.

[0010] As a further optimization scheme of the optimization method of the double-salient-pole memory motor of a humanoid robot described in the present invention, the optimization method of the double-salient-pole memory motor of a humanoid robot is specifically as follows:

[0011] Determine the constraints: Each stator slot space exceeds 80%;

[0012] Based on the constraints, the optimization satisfies the following magnetic load B e and torque density T v Index: B e =(p r +p s )(B H +B L )+(δ H +δ L )

[0013]

[0014] Among them, B H and BL are the magnetic flux density modulation of high coercivity permanent magnet and low coercivity permanent magnet, δ H and δ L are the fundamental harmonic components of high coercivity permanent magnet and low coercivity permanent magnet respectively, m represents the phase number of the double salient pole memory motor, F M and F L is the magnetomotive force amplitude of high coercivity permanent magnet and low coercivity permanent magnet, β is the air gap radius, k r and k s is the harmonic winding coefficient of the stator and rotor pole pairs, α is the number of modulated teeth, p s and P r are the stator and rotor pole pairs, ε H and ε L are the air gap permeability under the action of high coercive force permanent magnet and low coercive force permanent magnet respectively;

[0015] The objective function G(x) using Pareto weighting i ) min Quantify the sample database in the multi-attribute decision making of magnetic load and torque density indicators, so as to obtain the optimal solution of magnetic load and torque density indicators;

[0016] The Pareto weighted objective function is as follows:

[0017]

[0018] Among them, x i represents the i-th parameter variable in the sample space, B e (x i ), T v (x i ) represent the i-th optimized values ​​of magnetic load index and torque density index in the sample space, χ1 and χ2 represent the weight coefficients of magnetic load index and torque density index, respectively. Indicates the sample average value of magnetic load index and torque density index;

[0019] The Pearson coefficient r(·) is used to represent the linear relationship between the structural parameters of the double-salient memory motor and the magnetic load and torque density indicators.

[0020]

[0021] Where n is the number of sampling points.

[0022] As a further optimization scheme of the optimization method of the double-salient-pole memory motor of a humanoid robot described in the present invention,

[0023] Using the global optimal function r ci , optimize the correlation coefficients between the structural parameters of high coercivity permanent magnets and low coercivity permanent magnets and the magnetic load and torque density indicators;

[0024]

[0025] Among them, κ j is the weight coefficient of the jth optimization objective.

[0026] As a further optimization scheme of the optimization method of the double-salient-pole memory motor for a humanoid robot described in the present invention, a constraint condition is determined: the space of each stator slot exceeds 80%; specifically, as follows:

[0027]

[0028] Among them, S represents the constraint condition, Q N represents the stator slot full rate, B represents the stator slot area, ω m Indicates the coil coefficient, H c , L c Respectively represent the slot area occupied by each high-coercive force permanent magnet and low-coercive force permanent magnet.

[0029] As a further optimization scheme of the optimization method of the humanoid robot double-salient-pole memory motor described in the present invention, sensitivity is introduced to evaluate the comprehensive influence of high-coercivity permanent magnets and low-coercivity permanent magnets on the optimization targets of magnetic load and torque density, as follows:

[0030] H j (X i ) is the global optimal function r ci The sensitivity index of the jth optimization objective, H j (X i ) is expressed as:

[0031]

[0032] in, represents the average value of the global optimal function;

[0033] Introducing sensitivity G(X i ) is used to evaluate the comprehensive effect of high coercivity permanent magnets and low coercivity permanent magnets on the optimization objectives of magnetic load and torque density. i ) is expressed as:

[0034] G(X i )=λ1|H(B e (x i ))|+λ2|H(T v (x i ))|

[0035] Among them, H(B e (x i ))、H(T v (xi )) represent the sensitivity indexes of the structural parameters of high coercive force permanent magnet and low coercive force permanent magnet to the magnetic load and torque density, respectively. λ1 and λ2 are the pulsation coefficients of magnetic load and torque density, respectively.

[0036] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0037] (1) Compared with the optimization method of traditional memory motors, the parameter sensitivity multi-objective optimization design method of the present invention can improve the magnetic field anti-interference ability of the double-pole memory motor of the humanoid robot joint, significantly reduce the double-pole memory motor under rated conditions, can work in the magnetic field enhancement area, reduce the demagnetization risk of high coercive force permanent magnets and low coercive force permanent magnets, and can improve the speed regulation range of the motor;

[0038] (2) Compared with the response surface optimization method of the background technology, this method refines the design performance indicators, so that the coupling between the two permanent magnets of the double-pole memory motor of the humanoid robot joint is reduced, and the obtained structural parameters can converge to within a limited time, which can significantly improve the system response speed and greatly reduce the overshoot, and effectively improve the sinusoidality of the no-load back electromotive force of the memory motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the process of the present invention.

[0040] Figure 2 This is a no-load back electromotive force result diagram obtained by using the optimization method of the present invention.

[0041] Figure 3 This is a sensitivity analysis speed result diagram obtained by using the optimization method of the present invention.

[0042] Figure 4 This is a sensitivity analysis torque result diagram obtained by using the optimization method of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] The high coercivity permanent magnet, low coercivity permanent magnet and winding of the humanoid robot double-pole memory motor are all on the stator side. The mutual interference of the two permanent magnets leads to the complex magnetic field distribution of the stator structure and the tight space in the stator slot. The stator slot area needs to be optimized. To this end, starting from the requirements of meeting the magnetic load and torque density performance indicators of the double-salient pole memory motor, given the magnetic load and torque density indicators, a Pareto-weighted objective function is proposed to quantify the sample database in the multi-attribute decision of the two indicators, so as to obtain the optimal solution. The linear relationship between the structural parameters of the double-salient pole memory motor and the optimized performance indicators is measured by the Pearson coefficient. The sensitivity is introduced to evaluate the comprehensive influence of the two permanent magnets on the optimization objectives of magnetic load and torque density. Finally, the no-load back-electromotive force of the double-salient pole memory motor after optimization is better than the no-load back-electromotive force before optimization.

[0045] Figure 1 It is a flow chart of the present invention, given the two indicators of the double-salient-pole memory motor magnetic load and torque density, a Pareto-weighted objective function is used to quantify the sample database in the multi-attribute decision making of the two indicators of the double-salient-pole memory motor magnetic load and torque density, so as to obtain the optimal solution of the magnetic load and torque density indicators;

[0046] The linear relationship between the structural parameters of the double-pole memory motor and the optimal solution of the magnetic load and torque density indicators is measured by the Pearson coefficient. Through this linear relationship, the motor magnetic flux intensity and torque are optimized to be adjusted in real time with load changes.

[0047] Sensitivity is introduced to evaluate the combined effects of high-coercivity permanent magnets and low-coercivity permanent magnets on the optimization objectives of magnetic load and torque density.

[0048] Firstly, considering the limitations of the processing technology of high coercive force permanent magnets and low coercive force permanent magnets, it is determined that the space of each stator slot exceeds 80%.

[0049]

[0050] Among them, S represents the constraint condition, Q N represents the stator slot full rate, B represents the stator slot area, ω m Indicates the coil coefficient, H c , L c Respectively represent the slot area occupied by each high-coercive force permanent magnet and low-coercive force permanent magnet.

[0051] In order to improve the comprehensive performance of the doubly salient memory motor and to achieve the goal of improving the magnetic load and torque density, based on the constraints of formula (1), the magnetic load B is optimized to meet the following requirements: e and torque density T v Index: B e =(p r +p s)(B H +B L )+(δ H +δ L ) (2)

[0052]

[0053] Among them, B H and B L are the magnetic flux density modulation of high coercivity permanent magnet and low coercivity permanent magnet, δ H and δ L are the fundamental harmonic components of high coercivity permanent magnet and low coercivity permanent magnet respectively, m represents the phase number of the double salient pole memory motor, F M and F L is the magnetomotive force amplitude of high coercivity permanent magnet and low coercivity permanent magnet, β is the air gap radius, k r and k s is the harmonic winding coefficient of the stator and rotor pole pairs, α is the number of modulated teeth, p s and P r are the stator and rotor pole pairs, ε H and ε L They are the air gap permeability under the action of high coercive force permanent magnet and low coercive force permanent magnet respectively.

[0054] In order to quantify the sample database in the multi-attribute decision making of magnetic load and torque density indicators, a Pareto-weighted objective function G(x i ) min , thus obtaining the optimal solution, which is expressed as follows:

[0055]

[0056] Among them, x i represents the i-th parameter variable in the sample space, B e (x i ), T v (x i ) represent the i-th optimized values ​​of magnetic load index and torque density index in the sample space, χ1 and χ2 represent the weight coefficients of magnetic load index and torque density index, respectively. Indicates the sample average value of magnetic load index and torque density index.

[0057] Due to the correlation coefficient between the slot area occupied by high coercivity permanent magnets and low coercivity permanent magnets and the optimization target of magnetic load and torque density parameters, the Pearson coefficient r(·) is used here to represent the linear relationship between the structural parameters of the double-salient memory motor and the magnetic load and torque density indicators.

[0058]

[0059] In the formula, n is the number of sampling points.

[0060] Among them, in order to optimize the correlation coefficients between the structural parameters of high coercive force permanent magnets and low coercive force permanent magnets and the magnetic load and torque density indicators, the global optimal function r is used. ci ,

[0061]

[0062] In the formula, κ j is the weight coefficient of the jth optimization objective, H j (X i ) is the sensitivity index of the jth optimization objective in the global optimal function of formula (6), expressed as:

[0063]

[0064] In the formula, represents the average value of the global optimal function. In order to comprehensively evaluate the comprehensive influence of high coercivity permanent magnets and low coercivity permanent magnets on the optimization objectives of magnetic load and torque density, the sensitivity G(X i ), expressed as:

[0065] G(X i )=λ1|H(B e (x i ))|+λ2|H(T v (x i ))| (8)

[0066] Among them, H(B e (x i ))、H(T v (x i )) represent the sensitivity indexes of the structural parameters of high coercive force permanent magnet and low coercive force permanent magnet to the magnetic load and torque density, respectively. λ1 and λ2 are the pulsation coefficients of magnetic load and torque density, respectively.

[0067] According to the optimization method of formula (1-8), the sensitivity analysis results are as follows Figure 2 , Figure 3 , Figure 4 As shown, Figure 2 This is a sensitivity analysis no-load back electromotive force result diagram obtained by using the optimization method of the present invention. Figure 3 This is a sensitivity analysis speed result diagram obtained by using the optimization method of the present invention. Figure 4 The sensitivity analysis torque result diagram obtained by using the optimization method of the present invention shows that the no-load back electromotive force of the double-pole memory motor after optimization has better sinusoidal performance, higher peak value, better speed regulation performance, and stable increase in torque than the no-load back electromotive force before optimization.

[0068] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing a double-salient-pole memory motor for a humanoid robot, characterized in that: Given the two indicators of doubly salient memory motor magnetic load and torque density, a Pareto-weighted objective function is used to quantify the sample database in multi-attribute decision making of the two indicators of doubly salient memory motor magnetic load and torque density, so as to obtain the optimal solutions of the two indicators. The Pearson coefficient is used to measure the linear relationship between the structural parameters of the double-pole memory motor and the optimal solution of the magnetic load and torque density indicators. This linear relationship is used to optimize the motor's magnetic flux intensity and real-time adjustment of the torque as the load changes.

2. The method for optimizing a humanoid robot double-salient-pole memory motor according to claim 1, characterized in that: Also includes: Sensitivity is introduced to evaluate the combined effects of high-coercivity permanent magnets and low-coercivity permanent magnets on the optimization objectives of magnetic load and torque density.

3. The method for optimizing a humanoid robot double-salient-pole memory motor according to claim 1, characterized in that: The optimization method of the double-salient-pole memory motor of a humanoid robot is as follows: Determine the constraints: Each stator slot space exceeds 80%; Based on the constraints, the optimization satisfies the following magnetic load B e and torque density T v index: B e =(p r +p s )(B H +B L )+(d H +d L ) Among them, B H and B L are the magnetic flux density modulation of high coercivity permanent magnet and low coercivity permanent magnet, δ H and δ L are the fundamental harmonic components of high coercivity permanent magnet and low coercivity permanent magnet respectively, m represents the phase number of the double salient pole memory motor, F M and F L is the magnetomotive force amplitude of high coercivity permanent magnet and low coercivity permanent magnet, β is the air gap radius, k r and k s is the harmonic winding coefficient of the stator and rotor pole pairs, α is the number of modulated teeth, p s and P r are the stator and rotor pole pairs, ε H and ε L are the air gap permeability under the action of high coercive force permanent magnet and low coercive force permanent magnet respectively; The objective function G(x) using Pareto weighting i ) min Quantify the sample database in the multi-attribute decision making of magnetic load and torque density indicators, so as to obtain the optimal solution of magnetic load and torque density indicators; The Pareto weighted objective function is as follows: Among them, x i represents the i-th parameter variable in the sample space, B e (x i ), T v (x i ) represent the i-th optimized values ​​of magnetic load index and torque density index in the sample space, χ1 and χ2 represent the weight coefficients of magnetic load index and torque density index, respectively. Indicates the sample average value of magnetic load index and torque density index; The Pearson coefficient r(·) is used to represent the linear relationship between the structural parameters of the double-salient memory motor and the magnetic load and torque density indicators. Where n is the number of sampling points.

4. The method for optimizing a double-salient-pole memory motor of a humanoid robot according to claim 3, characterized in that: Using the global optimal function r ci , optimize the correlation coefficients between the structural parameters of high coercivity permanent magnets and low coercivity permanent magnets and the magnetic load and torque density indicators; Among them, κ j is the weight coefficient of the jth optimization objective.

5. The method for optimizing a double-salient-pole memory motor for a humanoid robot according to claim 3, characterized in that: Determine the constraints: each stator slot space exceeds 80%; the details are as follows: Among them, S represents the constraint condition, Q N represents the stator slot full rate, B represents the stator slot area, ω m Indicates the coil coefficient, H c , L c Respectively represent the slot area occupied by each high-coercive force permanent magnet and low-coercive force permanent magnet.

6. The method for optimizing a humanoid robot double-salient-pole memory motor according to claim 2, characterized in that: Sensitivity is introduced to evaluate the comprehensive impact of high coercivity permanent magnets and low coercivity permanent magnets on the optimization targets of magnetic load and torque density, as follows: H j (X i ) is the global optimal function r ci The sensitivity index of the jth optimization objective, H j (X i ) is expressed as: in, represents the average value of the global optimal function; Introducing sensitivity G(X i ) is used to evaluate the comprehensive effect of high coercivity permanent magnets and low coercivity permanent magnets on the optimization objectives of magnetic load and torque density. i ) is expressed as: G(X i )=λ1|H(B e (x i ))|+λ2|H(T v (x i ))| Among them, H(B e (x i ))、H(T v (x i )) represent the sensitivity indexes of the structural parameters of high coercive force permanent magnet and low coercive force permanent magnet to the magnetic load and torque density, respectively. λ1 and λ2 are the pulsation coefficients of magnetic load and torque density, respectively.

Images