Torque compensation method, air conditioner, device and system

Abstract

This disclosure relates to a torque compensation method, air conditioner, device, and system, specifically in the field of household appliances. It includes acquiring operating characteristic parameters of a motor, including an initial torque command and rotor electrical angle; performing complex domain mapping processing on the operating characteristic parameters to obtain a composite input vector; performing feature analysis processing on the composite input vector using a complex neural network to generate a set of complex harmonic coefficients corresponding to the number of successive synthesis units. Based on the set of complex harmonic coefficients and the rotor electrical angle, the torque compensation amount is determined; using the torque compensation amount, the initial torque corresponding to the initial torque command is compensated to obtain the compensated target torque command, thereby improving the reliability of torque compensation.

Description

Technical Field

[0001] This disclosure relates to the field of household appliances, and more particularly to torque compensation methods, air conditioners, equipment, and systems. Background Technology

[0002] Permanent magnet synchronous motors in variable frequency home appliances typically employ vector control, combined with harmonic injection, angle lookup table, or adaptive algorithms for torque compensation.

[0003] However, when the operating conditions change, the compensation phase and the rotor electrical angle are prone to become out of sync, resulting in a decrease in compensation accuracy. At the same time, the selection of the order and parameter tuning depend on experience, and some compensation links are coupled with the current closed loop, which can easily increase the difficulty of switching control and affect the stability of the closed loop, resulting in low reliability of torque compensation. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a torque compensation method, an air conditioner, an equipment, and a system.

[0005] According to a first aspect of the present disclosure, a torque compensation method is provided, comprising:

[0006] Collect the motor's operating characteristic parameters, including the initial torque command and rotor electrical angle;

[0007] The operational feature parameters are subjected to complex domain mapping to obtain a composite input vector;

[0008] The composite input vector is processed by a complex neural network to generate a set of complex harmonic coefficients that correspond to the number of synthesis units at each order.

[0009] The torque compensation amount is determined based on the set of complex harmonic coefficients and the rotor electrical angle.

[0010] The initial torque corresponding to the initial torque command is compensated by the torque compensation amount to obtain the compensated target torque command.

[0011] As an optional implementation, the operating characteristic parameters further include the quadrature-axis current feedback; the composite input vector is processed by a complex neural network to generate a set of complex harmonic coefficients corresponding to the number of successive synthesis units, including:

[0012] The estimated electromagnetic torque is determined based on the preset torque conversion factor and the cross-axis current feedback.

[0013] Determine whether the initial torque command and the estimated electromagnetic torque satisfy the enable condition;

[0014] If so, the complex input vector is analyzed and processed through the complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0015] As an optional implementation, determining whether the initial torque command and the estimated electromagnetic torque satisfy the enable condition includes:

[0016] Determine the torque error between the indicated electromagnetic torque corresponding to the initial torque command and the estimated electromagnetic torque;

[0017] If the absolute value of the torque error is greater than or equal to the torque error threshold, then the enabling condition is satisfied;

[0018] If the absolute value of the torque error is less than the torque error threshold, then the enabling condition is not met.

[0019] As an optional implementation, if the absolute value of the torque error is greater than or equal to the torque error threshold, then the enabling condition is satisfied, including:

[0020] If the absolute value of the torque error is greater than or equal to the torque error threshold, then update the number of consecutive cycles;

[0021] When the number of consecutive cycles is greater than or equal to the preset number of confirmation cycles, the enabling condition is determined to be met.

[0022] As an optional implementation, the complex neural network includes multiple layers of fully connected complex layers, a discrete complex activation function, and a complex coefficient output header; the complex neural network analyzes and processes the composite input vector to generate a set of complex harmonic coefficients corresponding to the number of successive synthesis units, including:

[0023] The composite input vector is transformed layer by layer using the multi-layer complex fully connected layer to obtain complex latent variable components.

[0024] The complex latent variable components are processed by the separated complex activation function to obtain the complex latent variables;

[0025] By using the complex coefficient output head, the complex latent variables are grouped and output in order to obtain a set of complex harmonic coefficients that corresponds to the number of the order synthesis units in order.

[0026] As an optional implementation, the operating characteristic parameters include torque command, rotor electrical angle, speed, direct-axis current feedback, and quadrature-axis current feedback; the operating characteristic parameters are subjected to complex domain mapping processing to obtain a composite input vector, including:

[0027] Determine the cosine component and sine component of the electrical angle corresponding to the rotor electrical angle;

[0028] The input feature vector is determined based on the torque command, the rotational speed, the DC current feedback, the quadrature axis current feedback, the cosine component of the electrical angle, and the sine component of the electrical angle.

[0029] Based on paired channels, the input feature vector is mapped to the complex field to obtain the composite input vector.

[0030] As an optional implementation, the complex harmonic coefficient set includes multiple complex harmonic coefficients, the number of which is equal to the number of successive synthesis units; the torque compensation amount is determined based on the complex harmonic coefficient set and the rotor electrical angle, including:

[0031] Based on the complex harmonic coefficients corresponding to each preset order and the rotor electrical angle, determine the step-by-step compensation amount corresponding to each preset order;

[0032] The sum of the successive compensation amounts corresponding to each preset order is determined as the torque compensation amount.

[0033] As an optional implementation, the complex harmonic coefficients include the real part and the imaginary part of the complex harmonics; for any preset order; based on the complex harmonic coefficients corresponding to the preset order and the rotor electrical angle, the successive compensation amount corresponding to the preset order is determined, including:

[0034] Based on the preset order and the rotor electrical angle, determine the order rotor cosine value and the order rotor sine value;

[0035] Determine the first product of the real part of the complex harmonic and the order rotor cosine value;

[0036] Determine the second product of the imaginary part of the complex harmonic and the order rotor sine value;

[0037] The difference between the first product and the second product is determined as the successive compensation amount.

[0038] As an optional implementation, the initial torque corresponding to the initial torque command is compensated using the torque compensation amount to obtain the compensated target torque command, including:

[0039] The sum of the torque compensation amount and the initial torque corresponding to the initial torque command is determined as the intermediate torque;

[0040] Based on the preset torque limit, the intermediate torque is limited to obtain the target torque corresponding to the target torque command;

[0041] The target torque command is generated based on the target torque.

[0042] As an optional implementation, generating the target torque command based on the target torque includes:

[0043] Determine the torque conversion factor of the motor;

[0044] When the torque conversion factor is greater than or equal to the lower threshold, the target torque is divided by the torque conversion factor to obtain the target quadrature axis current;

[0045] When the torque conversion factor is less than the lower threshold, the target torque is divided by the lower threshold to obtain the target quadrature axis current;

[0046] Generate the target torque command corresponding to the target quadrature axis current.

[0047] As an optional implementation, determining the torque conversion factor of the motor includes:

[0048] Obtain the direct-axis inductance parameters, quadrature-axis inductance parameters, permanent magnet flux linkage parameters, number of pole pairs, and direct-axis current feedback of the motor;

[0049] The inductance difference coupling term is determined based on the difference between the direct-axis inductance parameters and the quadrature-axis inductance parameters and the direct-axis current feedback.

[0050] The sum of the inductance differential coupling term and the permanent magnet flux linkage parameter is determined as the equivalent flux linkage term;

[0051] The product of the preset coefficient, the number of pole pairs, and the equivalent flux linkage term is determined as the torque conversion coefficient.

[0052] According to a second aspect of the present disclosure, a torque command compensation device includes a data acquisition module, a mapping module, an analysis module, a determination module, and a compensation module:

[0053] The acquisition module is used to acquire the operating characteristic parameters of the motor, including the initial torque command and the rotor electrical angle.

[0054] The mapping module is used to perform complex domain mapping processing on the running feature parameters to obtain a composite input vector.

[0055] The analysis module is used to perform feature analysis processing on the composite input vector through a complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0056] The determining module is used to determine the torque compensation amount based on the complex harmonic coefficient set and the rotor electrical angle.

[0057] The compensation module is used to compensate the initial torque corresponding to the initial torque command by using the torque compensation amount, so as to obtain the compensated target torque command.

[0058] As an optional implementation, the operating characteristic parameters further include quadrature-axis current feedback; the analysis module is specifically used for:

[0059] The estimated electromagnetic torque is determined based on the preset torque conversion factor and the cross-axis current feedback.

[0060] Determine whether the initial torque command and the estimated electromagnetic torque satisfy the enable condition;

[0061] If so, the complex input vector is analyzed and processed through the complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0062] As an optional implementation, the analysis module is specifically used for:

[0063] Determine the torque error between the indicated electromagnetic torque corresponding to the initial torque command and the estimated electromagnetic torque;

[0064] If the absolute value of the torque error is greater than or equal to the torque error threshold, then the enabling condition is satisfied;

[0065] If the absolute value of the torque error is less than the torque error threshold, then the enabling condition is not met.

[0066] As an optional implementation, the analysis module is specifically used for:

[0067] If the absolute value of the torque error is greater than or equal to the torque error threshold, then update the number of consecutive cycles;

[0068] When the number of consecutive cycles is greater than or equal to the preset number of confirmation cycles, the enabling condition is determined to be met.

[0069] As an optional implementation, the complex neural network includes multiple layers of fully connected complex layers, a split complex activation function, and a complex coefficient output header; the analysis module is specifically used for:

[0070] The composite input vector is transformed layer by layer using the multi-layer complex fully connected layer to obtain complex latent variable components.

[0071] The complex latent variable components are processed by the separated complex activation function to obtain the complex latent variables;

[0072] By using the complex coefficient output head, the complex latent variables are grouped and output in order to obtain a set of complex harmonic coefficients that corresponds to the number of the order synthesis units in order.

[0073] As an optional implementation, the operating characteristic parameters include torque command, rotor electrical angle, speed, direct-axis current feedback, and quadrature-axis current feedback; the mapping module is specifically used for:

[0074] Determine the cosine component and sine component of the electrical angle corresponding to the rotor electrical angle;

[0075] The input feature vector is determined based on the torque command, the rotational speed, the DC current feedback, the quadrature axis current feedback, the cosine component of the electrical angle, and the sine component of the electrical angle.

[0076] Based on paired channels, the input feature vector is mapped to the complex field to obtain the composite input vector.

[0077] As an optional implementation, the complex harmonic coefficient set includes multiple complex harmonic coefficients, the number of which is equal to the number of successive synthesis units; the determining module is specifically used for:

[0078] Based on the complex harmonic coefficients corresponding to each preset order and the rotor electrical angle, determine the step-by-step compensation amount corresponding to each preset order;

[0079] The sum of the successive compensation amounts corresponding to each preset order is determined as the torque compensation amount.

[0080] As an optional implementation, the complex harmonic coefficients include the real part and the imaginary part of the complex harmonics; for any preset order; the determining module is specifically used for:

[0081] Based on the preset order and the rotor electrical angle, determine the order rotor cosine value and the order rotor sine value;

[0082] Determine the first product of the real part of the complex harmonic and the order rotor cosine value;

[0083] Determine the second product of the imaginary part of the complex harmonic and the order rotor sine value;

[0084] The difference between the first product and the second product is determined as the successive compensation amount.

[0085] As an optional implementation, the compensation module is specifically used for:

[0086] The sum of the torque compensation amount and the initial torque corresponding to the initial torque command is determined as the intermediate torque;

[0087] Based on the preset torque limit, the intermediate torque is limited to obtain the target torque corresponding to the target torque command;

[0088] The target torque command is generated based on the target torque.

[0089] As an optional implementation, the compensation module is specifically used for:

[0090] Determine the torque conversion factor of the motor;

[0091] When the torque conversion factor is greater than or equal to the lower threshold, the target torque is divided by the torque conversion factor to obtain the target quadrature axis current;

[0092] When the torque conversion factor is less than the lower threshold, the target torque is divided by the lower threshold to obtain the target quadrature axis current;

[0093] Generate the target torque command corresponding to the target quadrature axis current.

[0094] As an optional implementation, the compensation module is specifically used for:

[0095] Obtain the direct-axis inductance parameters, quadrature-axis inductance parameters, permanent magnet flux linkage parameters, number of pole pairs, and direct-axis current feedback of the motor;

[0096] The inductance difference coupling term is determined based on the difference between the direct-axis inductance parameters and the quadrature-axis inductance parameters and the direct-axis current feedback.

[0097] The sum of the inductance differential coupling term and the permanent magnet flux linkage parameter is determined as the equivalent flux linkage term;

[0098] The product of the preset coefficient, the number of pole pairs, and the equivalent flux linkage term is determined as the torque conversion coefficient.

[0099] According to a third aspect of the present disclosure, an air conditioner is provided for performing the torque compensation method of any of the first aspects.

[0100] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising:

[0101] processor;

[0102] Memory used to store processor-executable instructions;

[0103] The processor is configured to implement the torque compensation method of any of the first aspects mentioned above.

[0104] According to a fifth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, comprising:

[0105] When the instructions in the storage medium are executed by the processor, the processor is able to perform any of the torque compensation methods described in the first aspect above.

[0106] According to a sixth aspect of the present disclosure, a computer program product is provided, comprising:

[0107] A computer program that, when executed by a processor, implements any of the torque compensation methods described in the first aspect above.

[0108] According to a sixth aspect of the present disclosure, a chip system is provided, comprising:

[0109] The chip system includes a processing unit and an interface circuit. The processing unit obtains program instructions through the interface circuit, and the program instructions are executed by the processing unit. The processing unit is used to execute the torque compensation method of any of the first aspects mentioned above.

[0110] The torque compensation method, air conditioner, device, and system provided in the embodiments of this disclosure can include the following beneficial effects: by collecting motor operating characteristic parameters including initial torque command and rotor electrical angle, and performing complex domain mapping processing on the operating characteristic parameters, a set of complex harmonic coefficients corresponding to the number of successive synthesis units is generated using a complex neural network. Then, the torque compensation amount is determined by combining the rotor electrical angle and the initial torque is compensated. This can simultaneously characterize the harmonic characteristics and their correspondence with the rotor electrical angle during the compensation process, improve the compensation phase synchronization and torque compensation accuracy under complex working conditions, and thus improve the stability and closed-loop controllability of the torque control of the permanent magnet synchronous motor.

[0111] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0112] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0113] Figure 1 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 1 ;

[0114] Figure 2 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 2 ;

[0115] Figure 3 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 3 ;

[0116] Figure 4 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 4 ;

[0117] Figure 5 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 5 ;

[0118] Figure 6 This is a schematic diagram of the structure of an air conditioning control device according to some embodiments of the present disclosure;

[0119] Figure 7 This is a block diagram illustrating an electronic device according to some embodiments of the present disclosure;

[0120] Figure 8 This is a block diagram illustrating another electronic device according to some embodiments disclosed in a book;

[0121] Figure 9 This is a block diagram of a chip system according to some embodiments of the present disclosure. Detailed Implementation

[0122] Some embodiments of this disclosure will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a particular order. Furthermore, for clarity and brevity, descriptions of features known in the art may be omitted.

[0123] The embodiments described in the following examples of this disclosure are not representative of all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0124] Torque compensation technology for permanent magnet synchronous motors is mainly used in the motor control field of variable frequency home appliances, such as air conditioner compressors, fans, and washing machine drive systems. These systems typically employ a vector control architecture, where the controller generates target control quantities based on torque commands, rotor position, and motor operating status to achieve coordinated control of speed, torque, and stability.

[0125] In the above application scenarios, the motor operation is easily affected by cogging effect, parameter drift and non-ideal characteristics of power devices, which in turn generate torque disturbance in the periodic domain related to rotor electrical angle, thus affecting the noise, vibration and energy efficiency of the whole machine.

[0126] In related technologies, permanent magnet synchronous motors in variable frequency home appliances typically employ vector control, combined with harmonic injection, angle lookup tables, or adaptive algorithms for torque compensation. However, when operating conditions change, the compensation phase and rotor electrical angle are prone to asynchrony, leading to a decrease in compensation accuracy. At the same time, the selection of the order and parameter tuning depend on experience, and some compensation links are coupled with the current closed loop, which can easily increase the difficulty of switching control and affect the stability of the closed loop, resulting in low reliability of torque compensation.

[0127] To address the aforementioned technical problems, this disclosure proposes the following concept: by collecting the operating characteristic parameters of the motor and performing complex domain mapping processing on the operating characteristic parameters to obtain a composite input vector, a set of complex harmonic coefficients corresponding to a preset number of orders is generated through a complex neural network, and the torque compensation amount is determined by combining the rotor electrical angle. Finally, the initial torque corresponding to the initial torque command is compensated to obtain the compensated target torque command, thereby improving the reliability of torque compensation.

[0128] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0129] Figure 1 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 1 ,like Figure 1 As shown, the torque compensation method is used in the motor controller of variable frequency home appliances and includes the following steps.

[0130] S101. Collect the operating characteristic parameters of the motor.

[0131] Operating characteristic parameters may include initial torque command and rotor electrical angle.

[0132] The initial torque command is used to characterize the target torque output of the motor and is the starting point for compensation in this method. The rotor electrical angle is used to provide the phase reference for torque compensation, ensuring that the compensation amount is synchronized with the motor rotor position.

[0133] The initial torque command can be output from the upper speed regulation loop or directly given by the overall machine control strategy based on the current operating conditions. When the control system adopts a vector control architecture, the initial torque command corresponds to the original target value before entering the torque channel.

[0134] The rotor electrical angle can be calculated from the detection results of the position sensor, which can be an encoder, a rotary transformer, or a Hall position detection unit.

[0135] In sensorless implementations, the rotor electrical angle can also be estimated by a back EMF observer, a flux linkage observer, or a phase-locked loop.

[0136] In practice, the initial torque command is read from the torque command register at the same sampling time, then the angle data output by the position sampling module is read, and an electrical angle conversion is performed on the angle data. The conversion relationship can be expressed as follows: ,in This indicates the rotor electrical angle, and p indicates the number of pole pairs. This represents the mechanical angle; if the estimated electrical angle is obtained directly, it can be directly used as the rotor electrical angle input.

[0137] For the collected parameters, normalization processing can also be performed. For example, the initial torque command can be normalized according to the rated torque standard, and the rotor electrical angle can be limited to the periodic range of 0 to 2π, or converted to the symmetrical range of -π to π.

[0138] Based on the above analysis, this step obtains the initial torque command and rotor electrical angle synchronously to form a set of operating characteristic parameters consistent with the current control state. This ensures that subsequent compensation calculations do not depend on asynchronous angles or lag state quantities, thus providing a clear phase reference for the compensation quantity generation process and reducing the misalignment between the compensation phase and the actual disturbance under complex operating conditions.

[0139] S102. Perform complex domain mapping on the running feature parameters to obtain a composite input vector.

[0140] Complex domain mapping is used to transform the collected running feature parameters into an input form suitable for processing by complex-valued neural networks (CVNNs), thereby preserving phase information and improving feature representation capabilities.

[0141] The composite input vector is used to carry the input features of the complex neural network. It is an intermediate data structure obtained by mapping the running feature parameters through the complex field.

[0142] In this application, after obtaining the set of running feature parameters, one or more complex input components are constructed according to a preset mapping rule. Each complex input component contains a real part and an imaginary part. Multiple complex input components are arranged in a predetermined order to form a composite input vector.

[0143] In practical implementation, the initial torque command and the information corresponding to the rotor electrical angle can be converted into complex domain expressions according to a preset method to maintain a uniform representation of the input data in the complex neural network. To ensure consistent network input scale, amplitude limiting and linear scaling can also be applied to each complex input component to ensure that the real and imaginary parts fall within a preset numerical range. After the composite input vector is formed, it is stored in the controller's internal storage area in vector form and fed into the complex neural network within the current control cycle.

[0144] Based on the above analysis, this step maps the running feature parameters to the complex domain expression, so that the input features carry the phase information corresponding to the rotor position. When the network learns periodic disturbances, it can use the amplitude and phase relationship in the complex form, thus providing the input basis for the generation of complex harmonic coefficients.

[0145] S103. The composite input vector is processed by a complex neural network to generate a set of complex harmonic coefficients that correspond to the number of synthesis units at each order.

[0146] Complex neural networks are used to perform feature analysis on composite input vectors and output a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0147] The set of complex harmonic coefficients is used to characterize the compensation information corresponding to each preset order and serves as the core basis for determining the torque compensation amount.

[0148] Complex neural networks are deployed in the software model layer of the controller or in a dedicated inference unit to extract and map features from composite input vectors according to complex number operation rules, thereby obtaining a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0149] The number of successive synthesis units corresponds to the set of harmonic orders to be compensated. For example, it can be set to the first to the Nth order, where N can be determined based on the motor structure characteristics, cogging effect spectrum, or control resources. When the preset order is k, the network outputs a set of corresponding complex harmonic coefficients. ,in Characterizing the in-phase component of this harmonic order, The orthogonal components characterize the harmonic order.

[0150] Within the current control cycle, the composite input vector is input into the network, and after forward inference, a set of complex harmonic coefficients is obtained. .

[0151] In one possible embodiment, the network parameters are pre-trained offline and stored in the controller memory. The training samples can consist of operating data under different speed, load and temperature rise conditions, and the target label is the complex representation of each order of torque disturbance.

[0152] In another implementation, online fine-tuning can also be performed during equipment operation.

[0153] This step utilizes a complex neural network to establish a mapping relationship between operating characteristics and harmonic compensation coefficients of each order. It no longer relies on fixed lookup table values ​​or empirically set single harmonic parameters. Furthermore, by simultaneously encoding amplitude and phase information through complex coefficients, the compensation parameters under different operating conditions can be updated synchronously with changes in the input state, thereby ensuring a higher consistency between the subsequently generated compensation amount and the actual periodic disturbance.

[0154] S104. Determine the torque compensation amount based on the set of complex harmonic coefficients and the rotor electrical angle.

[0155] Torque compensation is used to offset the error component in the initial torque of the motor, thereby correcting the target torque command.

[0156] The torque harmonic components are synthesized step by step using the set of complex harmonic coefficients and the rotor electrical angle of the current control cycle, and the total torque compensation is obtained by accumulating the components of each order.

[0157] In practical implementation, for the k-th order complex harmonic coefficient ck = ak + jbk, the compensation component corresponding to this order is determined using the rotor electrical angle θe as the phase reference. After performing the same processing on all preset orders, the torque compensation amount is obtained. If the complex coefficients output by the network have been standardized, they can be de-standardized and restored according to the rated torque after accumulation to obtain the compensation amount in the actual torque unit.

[0158] To limit abrupt changes in the compensation output, amplitude and rate of change constraints can be applied to the torque compensation amount. For example, the compensation amount can be limited to a preset proportion of the rated torque, and the slope of the compensation difference between adjacent control cycles can be limited.

[0159] This step does not directly use static compensation values ​​that are independent of angle. Instead, it uses the rotor electrical angle as a unified phase reference and determines the torque compensation amount based on the set of complex harmonic coefficients. Therefore, under conditions of speed and load changes, the compensation phase can still be kept to change synchronously with the rotor movement, reducing residual torque pulsation caused by compensation misalignment.

[0160] S105. By using torque compensation, the initial torque corresponding to the initial torque command is compensated to obtain the compensated target torque command.

[0161] The initial torque is used to represent the actual or corresponding original torque state of the motor before compensation, and is the object of compensation processing. The target torque command after compensation is used as the final torque target value output to the motor control link.

[0162] The torque compensation is superimposed on the torque channel corresponding to the initial torque command to obtain the compensated target torque command.

[0163] In practical implementation, according to the relational formula Generate the compensated target torque command, where Indicates the target torque command. This indicates the initial torque command. This represents the torque compensation amount. When the control system operates using standardized values, the above calculations are completed within the standardized domain and then converted into actual physical quantities before being output to subsequent modules.

[0164] To ensure that the target command is within the allowable range, the target torque command can be further restricted after superposition.

[0165] This step directly applies the obtained torque compensation to the link corresponding to the initial torque command, forming a target torque command that can be executed by the subsequent control module. This allows the compensation result to enter the existing control framework in the form of control target correction, thereby maintaining the consistency of closed-loop control.

[0166] The torque compensation method provided in this application can collect the operating characteristic parameters of a motor, including the initial torque command and the rotor electrical angle; perform complex domain mapping processing on the operating characteristic parameters to obtain a composite input vector; perform feature analysis processing on the composite input vector through a complex neural network to generate a set of complex harmonic coefficients corresponding to the number of successive synthesis units; determine the torque compensation amount based on the set of complex harmonic coefficients and the rotor electrical angle; and compensate the initial torque corresponding to the initial torque command using the torque compensation amount to obtain the compensated target torque command. In this application, the operating characteristic parameters are first organized into a composite input vector in complex form, then the complex neural network outputs successive complex harmonic coefficients, and then the torque compensation amount is determined based on the rotor electrical angle as the phase reference. Finally, the obtained torque compensation amount is applied to the torque command channel. This implementation method enables the amplitude and phase information of the compensation parameters to be continuously transmitted within the same data link, and keeps the compensation output synchronously corresponding to the rotor electrical angle. It can still generate the corresponding compensation amount around the current operating state under conditions of speed, load, and temperature rise changes.

[0167] It should be understood that the above examples are merely illustrative and not limiting. In one possible embodiment, the implementation of complex field mapping and the specific software deployment location can be adjusted accordingly without departing from the core technical ideas of this application.

[0168] Based on the above embodiments, the following, in conjunction with Figure 2 The execution process of performing complex domain mapping on the running feature parameters to obtain a composite input vector in S102 provided in this application is explained in detail.

[0169] Figure 2This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 2 ,like Figure 2 As shown, the torque compensation method is used in the motor controller of variable frequency home appliances and includes the following steps.

[0170] S201. Determine the cosine component of the electrical angle and the sine component of the electrical angle corresponding to the rotor electrical angle.

[0171] Operating characteristic parameters, except for the initial torque command Rotor electrical angle In addition, it also includes rotational speed. DC current feedback quantity Quadrature axis current feedback quantity .

[0172] The cosine component of the electrical angle is obtained by performing a cosine transform on the rotor electrical angle. The rotor electrical angle is then subjected to a sinusoidal transformation to obtain the sinusoidal component of the electrical angle. .

[0173] S202. Determine the input feature vector based on the torque command, speed, DC current feedback, quadrature axis current feedback, cosine component of electrical angle, and sine component of electrical angle.

[0174] Initial torque command Rotation speed DC current feedback quantity Quadrature axis current feedback quantity and the cosine component of the electrical angle Sine component of electrical angle The input feature vector is formed by concatenating the components in a preset order. The input feature vector is a six-dimensional real number feature.

[0175] To maintain consistency in the numerical range of inputs with different dimensions, the initial torque command... Rotation speed DC current feedback quantity Quadrature axis current feedback quantity Linear scaling is performed by applying preset dimensional coefficients, which are determined by the reciprocal of the corresponding rated value or the maximum allowable value pre-stored by the controller, and the cosine component of the electrical angle. Sine component of electrical angle Keep the original input values, so that the input feature vector It is consistent with the definition of the training input for complex neural networks.

[0176] S203. Based on paired channels, perform complex field mapping on the input feature vector to obtain a composite input vector.

[0177] According to the preset pairing rules for channels, the two adjacent real features of the input feature vector are taken as the real part and the imaginary part, respectively, and combined into a single complex input component. All complex input components are integrated to form a composite input vector with regular dimensions.

[0178] The complex field mapping logic of paired channels can be implemented by the controller's built-in software algorithm or by a dedicated embedded computing module hardware. This application does not limit the specific implementation carrier or module model.

[0179] Composite input vector For three-dimensional features, the complex field mapping applies a pairwise channel rule to the input feature vector. Convert to composite input vector Composite input vector The first real part is the input feature vector The first dimension, where the imaginary part of the first dimension is the input feature vector. The second dimension; composite input vector The second real part is the input feature vector The third dimension, the imaginary part of the second dimension is the input feature vector. The fourth dimension; composite input vector The real part of the third dimension is the input feature vector. The fifth dimension, where the imaginary part of the third dimension is the input feature vector. The sixth dimension.

[0180] For example, let the six-dimensional real-valued input feature vector be... After mapping through the complex field, a three-dimensional composite input vector is obtained. This completes the structured transformation from real number features to orthogonal complex number features.

[0181] The composite input vector mapped to the complex domain can fully preserve the coupling relationship between the multi-source features of torque, electrical angle, speed and current feedback, making the extraction and representation of the periodic disturbance features of the motor by the complex neural network more stable and consistent.

[0182] Since the direct-axis and quadrature-axis current feedback quantities are first converted into corresponding cosine and sine quadrature components, and then mapped into a standardized complex input structure through paired channel rules, the input features are more suitable for subsequent complex neural network processing in terms of phase correlation and numerical continuity, thus providing a stable input for generating complex harmonic coefficients corresponding to the perturbation order.

[0183] The torque compensation method provided in this application has its operating feature parameters restructured before entering the complex neural network, which can improve the synchronicity between the features and the rotor electrical angle, and enhance the consistency of the representation of the periodic torque disturbance of the motor, thereby providing a more reliable input basis for the subsequent calculation of torque compensation.

[0184] Based on the above embodiments, the following, in conjunction with Figure 3 The execution process of generating a set of complex harmonic coefficients corresponding to the number of successive synthesis units by performing feature analysis on the composite input vector through a complex neural network in S103 provided in this application is described in detail.

[0185] Figure 3 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 3 ,like Figure 3 As shown, the torque compensation method is used in the motor controller of variable frequency home appliances and includes the following steps.

[0186] S301. Determine the estimated electromagnetic torque based on the preset torque conversion coefficient and the quadrature axis current feedback.

[0187] Operating characteristic parameters also include quadrature axis current feedback. .

[0188] By combining the motor torque characteristics, a preset torque conversion coefficient is used to convert the quadrature axis current feedback in real time, resulting in an estimated electromagnetic torque with dimensions and amplitude scale that are consistent with the initial torque command. This estimation method can be adapted to real-time torque observation under steady-state and slight disturbance conditions of the motor, avoiding the defect that a single calibration constant cannot be adapted to dynamic conditions.

[0189] Example: Preset motor rated torque conversion factor To fix the calibration parameters, the quadrature-axis current feedback value of the current control cycle is collected. Through torque estimation formula The estimated electromagnetic torque is obtained by real-time solution, where The torque-current conversion factor is preset for the motor, and the final estimated electromagnetic torque is obtained. With dimensions consistent with the initial torque command, it can be directly used for subsequent torque deviation calculation and feature extraction.

[0190] S302. Determine whether the initial torque command and estimated electromagnetic torque meet the enable conditions.

[0191] The difference between the initial torque command and the estimated electromagnetic torque is judged, and the difference is compared with the preset enable threshold. When the difference is within the set range, the enable condition is deemed to be met.

[0192] In some embodiments, the torque error between the indicated electromagnetic torque corresponding to the initial torque command and the estimated electromagnetic torque is determined; if the absolute value of the torque error is greater than or equal to the torque error threshold, the enable condition is satisfied; if the absolute value of the torque error is less than the torque error threshold, the enable condition is not satisfied.

[0193] After receiving the initial torque command, the system first calculates the indicated electromagnetic torque that corresponds one-to-one with the command, then calls the estimation model based on the quadrature axis current feedback to output the estimated electromagnetic torque, and performs a difference calculation between the two to obtain the original torque error.

[0194] It is worth noting that the determination process requires the magnitude of the torque error, which is achieved by taking the absolute value. That is, the absolute value of the original torque error is directly obtained without distinguishing between positive and negative values ​​or assigning the inverse value.

[0195] The absolute value of the obtained torque error is compared with the torque error threshold. If the torque error amplitude is greater than or equal to the torque error threshold, a judgment signal that satisfies the compensation enable condition is output, allowing the complex neural network to run and participate in the generation of complex harmonic coefficients.

[0196] If the torque error amplitude is less than the torque error threshold, a judgment signal indicating that the compensation enable condition is not met will be output, the compensation link will remain locked, and the torque compensation operation will not be started.

[0197] In this embodiment, the compensation link is enabled only when the deviation between the initial torque command and the estimated electromagnetic torque reaches a preset threshold, so that the torque compensation action matches the strength of the actual disturbance. This enable determination is based solely on the torque error amplitude, which can avoid frequent triggering of invalid compensation under small error conditions and improve the stability and consistency of the overall compensation control.

[0198] In some embodiments, if the absolute value of the torque error is greater than or equal to the torque error threshold, the number of consecutive cycles is updated; when the number of consecutive cycles is greater than or equal to the preset confirmation cycle number, it is determined that the enable condition is met.

[0199] In each sampling period, the absolute value of the current torque error is compared with the torque error threshold. When the comparison result shows that the absolute value of the torque error is greater than or equal to the torque error threshold, the counter increments the continuous period count by one. When the comparison result shows that the absolute value of the torque error is less than the torque error threshold, the continuous period count is reset to zero or kept at a preset initial value so that continuity can be reconfirmed.

[0200] When the number of consecutive cycles reaches the preset number of confirmation cycles, the judgment result that the enable condition is met is output, and the subsequent complex neural network is allowed to output non-zero complex harmonic coefficients; when the number of consecutive cycles does not reach the preset number of confirmation cycles, the state of not meeting the enable condition is maintained, and the complex neural network outputs zero complex harmonic coefficients, thereby making the torque compensation amount zero.

[0201] During operation, by continuously and periodically confirming the over-threshold state, short-term disturbances are distinguished from stable deviations, and the compensation link is only activated after the continuous over-threshold reaches a set duration. This method matches the compensation enable determination with the persistence of torque error, reduces false triggering caused by unstable fluctuations, and ensures consistency between the generation and deactivation of subsequent torque compensation amounts.

[0202] In this embodiment, the compensation enable condition is constrained by the number of consecutive cycles, and the compensation output will not switch frequently due to the fluctuation of instantaneous error within a single cycle. The output state of the complex neural network is more stable, and the start and stop of the torque compensation is smoother, thereby improving the controllability and operational stability of the torque compensation link.

[0203] S303. If so, the complex input vector is analyzed and processed through a complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0204] Once the enabling condition is met, the composite input vector obtained from the previous complex domain mapping can be input into the complex neural network. The network extracts the periodic features related to the rotor electrical angle according to the complex weight parameters of each order, and outputs a set of complex harmonic coefficients corresponding to the number of synthesis units of each order.

[0205] Complex neural networks consist of multiple layers of fully connected complex layers, a split complex activation function, and a complex coefficient output head.

[0206] In some embodiments, a multi-layer complex fully connected layer can be used to transform the composite input vector layer by layer to obtain complex latent variable components; the complex latent variable components are processed by a separate complex activation function to obtain complex latent variables; the complex latent variables are output in order-wise groups through a complex coefficient output head to obtain a set of complex harmonic coefficients corresponding to the number of synthesis units in order.

[0207] Composite input vector It can be determined by the initial torque command and the rotor electrical angle. The speed, DC current feedback quantity, and quadrature axis current feedback quantity are obtained by mapping in the complex domain. The mapped data of each dimension are used as the real part and imaginary part respectively as inputs to the multilayer complex fully connected layer.

[0208] Multilayer complex fully connected layers combine input vectors Perform layer-by-layer transformation to generate complex latent variable components .

[0209] A multi-layer complex fully connected layer consists of at least two complex fully connected layers. Each complex fully connected layer contains several complex neurons and uses complex parameters for weighted summation. Real part computation pathways and imaginary part computation pathways are executed in each complex fully connected layer to obtain complex latent variable components. A split complex activation function is applied to the complex latent variable components to obtain complex latent variables.

[0210] The complex parameters of the fully connected complex layer are obtained through offline training and are stored in a parameter table that can be read by the controller. When the controller is powered on and initialized, it loads the real part weight matrix, imaginary part weight matrix, real part bias vector and imaginary part bias vector of each layer from the parameter table and calls them according to the control cycle during inference. Parameter updates are not performed within the control cycle.

[0211] The single-variable nonlinear transformation used in the separable complex activation function In this embodiment, hyperbolic tangent transformation is used and implemented by lookup table plus linear interpolation.

[0212] Specifically, using the input scalar as the index, adjacent values ​​are read from a preset hyperbolic tangent table and linearly interpolated to output a univariate nonlinear transformation. The result is that the univariate nonlinear transformation Having a defined numerical generation process and capable of real-time computation on the controller, to limit the computational paths of the real and imaginary parts of the layer-by-layer transformation and the separated complex activation function to directly implementable computational links, the complex eigentransform layer is in the first... The transformation of a layer onto a complex vector satisfies the following equation:

[0213] ;

[0214] ;

[0215] ;

[0216] ;

[0217] ;

[0218] in, The input complex vector is the complex feature transform layer. For a composite input vector, For the first The output complex vector of a complex fully connected layer. for The real part vector, for The imaginary part vector, For the first The real part of the complex weight matrix of the layer. For the first The real part of the complex weight matrix of the layer. For the first The imaginary part of the layer complex weights, Let be the real part bias vector of the first layer complex bias. For the first The imaginary part bias vector of the layer complex bias. For the first The intermediate vector of the real part of the layer. This is the intermediate vector of the imaginary part of the first layer. The hyperbolic tangent transform used for the separable complex activation function. The imaginary unit, L is the layer index, and L is the number of complex fully connected layers. These are latent variables for complex numbers.

[0219] The split complex activation function directly acts on the complex intermediate quantities output by each layer, and after nonlinear transformation, it forms complex hidden variables, which are then sent to the complex coefficient output head.

[0220] The complex coefficient output head is internally configured with independent complex output branches corresponding to preset orders. Each branch outputs only the complex harmonic coefficients of the corresponding order, thus obtaining a set of complex harmonic coefficients.

[0221] The preset order is taken entirely from a finite set of orders. finite order set It contains all target harmonic orders to be solved, and all elements in the set are the preset orders, and the two are completely equivalent in range.

[0222] For sets of finite order Within the preset order range ( Generate a finite set of orders according to the order increment rule (where the upper limit threshold is the order limit). .

[0223] The order-increasing rule is to start from order 1 and increase sequentially to the highest order. And write all the integer orders obtained from the traversal into a finite order set in sequence. At the same time, the set of finite orders The actual number of members is assigned as the number of successive synthesis units. , making a set of finite order The parameter adjustment can synchronously trigger the adaptive change of the overall scale of the network's step-by-step output structure and harmonic synthesis path, without the need for independent configuration and individual parameter debugging of a single harmonic order.

[0224] Latent variables of complex numbers Input complex coefficients and output headers, and establish a finite-order set using a successive grouping output structure. The complex output branches are aligned stepwise, and the complex coefficient output headers are based on a finite set of orders. The order of each order is Set up a separate, independent branch for complex number output.

[0225] Each complex output branch contains at least one fully connected complex layer to extract data from complex latent variables. Extract the complex harmonic coefficients specific to this order. .

[0226] Each complex output branch is bound to the corresponding order of the complex harmonic coefficient storage bits within the complex harmonic coefficient interface. This binding relationship consists of a finite set of orders. The only certainty is that the number of output channels remains constant throughout the entire operation; thus, the number of output channels and the finite set of orders are realized. The number of elements is equal, and different orders occupy dedicated storage spaces independently. There is no channel reuse or data aliasing of coefficients of different orders, and no cross-order mixed coefficients are generated.

[0227] Each complex output branch outputs complex harmonic coefficients in turn. And directly write to the complex harmonic coefficient interface, one group It can simultaneously carry both amplitude and phase information of the corresponding order harmonics.

[0228] by The real and imaginary parts serve as the two output components of the complex harmonic coefficients of that order, and can match the input format of the complex harmonic coefficient interface without additional phase reconstruction and component recombination operations; the output stage follows a finite order set. The order of each element is sequentially filled into the corresponding storage bits, ensuring that each order of the periodic base waveform in the electrical angle synchronous complex harmonic synthesizer can receive a corresponding complex driving quantity.

[0229] It is worth noting that the above network structure can also adopt other numbers of layers, output channels, or activation methods, and this application does not limit these.

[0230] During operation, the composite input vector is first mapped to a complex representation. Then, the implicit features related to the periodic disturbance are extracted through a multi-layer complex fully connected layer. After the nonlinear expression is enhanced by a separate complex activation function, the complex coefficient output head outputs the complex harmonic coefficients of each order in turn. Finally, it ensures that each order of compensation component corresponds one-to-one with the preset order and can be directly used for subsequent torque compensation synthesis.

[0231] In this embodiment, a complex coefficient structure with independent output at each order is adopted, so that the compensation information of each order will not be mixed. The correspondence between the complex harmonic coefficient and the rotor electrical angle is more stable, and it is easier to accumulate by order when output to the compensation link. This makes the composition of the torque compensation amount clearer and the order control more consistent, thereby improving the compensation accuracy and control stability under complex working conditions.

[0232] In some embodiments, when the compensation enable condition is not met, the output of the complex neural network is frozen or set to zero to avoid invalid compensation amounts from participating in subsequent calculations.

[0233] It can output zero complex harmonic coefficients corresponding one-to-one with their order to the complex harmonic coefficient interface, and restrict the generation of non-zero complex harmonic coefficients by the order-by-order grouping output structure of the complex coefficient output head.

[0234] The zero complex output path will handle each order The corresponding zeroth harmonic coefficient Assign a complex number with the value of zero, and follow the set of finite order. The sequence is written to the corresponding storage bits, so that the electrical angle synchronous complex harmonic synthesizer inputs zero torque compensation when synthesizing the waveform step by step.

[0235] At the same time, gating logic is used to block complex hidden variables. The input is fed into the complex coefficient output head, or the internal output register of the complex coefficient output head is directly overwritten to ensure that the complex harmonic coefficient interface has no non-zero value output when the disabled condition is not enabled.

[0236] The torque compensation method provided in this application incorporates the quadrature axis current feedback into the operating characteristic parameters and uses the matching relationship between the initial torque command and the estimated electromagnetic torque to control the output of the complex neural network. This allows the harmonic coefficients to be generated only under effective operating conditions, thereby improving the stability and scenario adaptability of the coefficient output and reducing the interference of invalid compensation on the subsequent torque compensation calculation.

[0237] Based on the above embodiments, the following, in conjunction with Figure 4 The execution process of determining the torque compensation amount based on the set of complex harmonic coefficients and the rotor electrical angle in S104 provided in this application is described in detail. The set of complex harmonic coefficients includes multiple complex harmonic coefficients, and the number of these multiple complex harmonic coefficients is the number of successive synthesis units.

[0238] Figure 4 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 4 ,like Figure 4 As shown, the torque compensation method is used in the motor controller of variable frequency home appliances and includes the following steps.

[0239] S401. Determine the step-by-step compensation amount corresponding to each preset order based on the complex harmonic coefficients and rotor electrical angles corresponding to each preset order.

[0240] Each complex harmonic coefficient can include the real part and the imaginary part of the complex harmonic. After reading the rotor electrical angle, the rotor electrical angle is converted into the corresponding complex exponential quantity, and then multiplied with the complex harmonic coefficient of each order to obtain the order-by-order compensation quantity.

[0241] In some embodiments, for any preset order, the order rotor cosine value and the order rotor sine value are determined according to the preset order and the rotor electrical angle; a first product of the real part of the complex harmonic and the order rotor cosine value is determined; a second product of the imaginary part of the complex harmonic and the order rotor sine value is determined; and the difference between the first product and the second product is determined as the order-by-order compensation amount.

[0242] For a preset order, the rotor electrical angle is subjected to frequency doubling, and cosine and sine function calculations are performed separately to obtain the order rotor cosine and sine values. Subsequently, the real part of the complex harmonics output by the complex neural network is multiplied by the order rotor cosine value, and the imaginary part of the complex harmonics is multiplied by the order rotor sine value. The difference between the two is then calculated to obtain the order-wise compensation amount. If multiple preset orders exist, the order-wise compensation amount for each order can be calculated separately in the same way and then synthesized into the torque compensation amount in subsequent stages.

[0243] Regarding the first The successive compensation amounts for each order can be found in the following formula:

[0244]

[0245] in, For the first The order-wise compensation amount, For the first The real part of the complex harmonics of the order of the complex harmonic coefficients. For the first The complex harmonic imaginary part of the complex harmonic coefficients of order, For the first The order of the rotor cosine value, For the first The order of the rotor sine value, The rotor electrical angle.

[0246] It can be executed synchronously within the control cycle, and the corresponding array storage area can be configured according to the number of successive synthesis units to match the compensation requirements of different orders.

[0247] In this embodiment, by multiplying the real part of the complex harmonics with the order rotor cosine value and multiplying the imaginary part of the complex harmonics with the order rotor sine value and taking the difference, the step-by-step compensation amount can maintain a synchronous correspondence with the rotor electrical angle. This allows the step-by-step compensation amount to directly reflect the effective contribution of each order harmonic component at the current angle, thereby improving the consistency of the compensation expression and the angle matching accuracy, and providing a stable order basis for the generation of subsequent torque compensation commands.

[0248] S402. The sum of the step-by-step compensation amounts corresponding to each preset order is determined as the torque compensation amount.

[0249] Each order of compensation corresponds to the periodic torque disturbance components of the motor at different frequencies, and all of them use the rotor electrical angle as a unified phase reference to ensure that the phase of each order of compensation is aligned and there is no phase offset. The compensation quantities of each order are superimposed step by step under the same rotor angle reference, which can synthesize a torque compensation quantity that matches the current rotor position in real time and accurately cancels the periodic disturbance.

[0250] Please refer to the following formula:

[0251]

[0252] in, This is the torque compensation amount. For the first The order-wise compensation amount corresponding to each order. It is a set of finite order.

[0253] The torque compensation amount acts directly on the motor torque channel to correct the original torque command in real time. The compensation amount is synthesized by the stepwise compensation amount obtained by mapping the complex harmonic coefficients of each order. The superposition calculation is completed entirely based on the unified reference of the rotor electrical angle to ensure that the compensation calculation is completely synchronized with the real-time angle position of the motor.

[0254] The torque compensation method provided in this application embodiment can generate corresponding step-by-step compensation amounts according to a preset order of complex harmonic coefficient set, and complete the unified phase alignment by rotor electrical angle. Then, the torque compensation amount is obtained by step-by-step summation, so that the compensation result has the same periodic characteristics as the rotor position, thereby improving the correspondence of compensation calculation and output stability.

[0255] Based on the above embodiments, the following, in conjunction with Figure 5 The execution process of S105 provided in this application, which compensates the initial torque corresponding to the initial torque command by using torque compensation amount to obtain the compensated target torque command, is explained in detail.

[0256] Figure 5 This is a flowchart illustrating a torque compensation method according to some embodiments of the present disclosure. Figure 5 ,like Figure 5As shown, the torque compensation method is used in the motor controller of variable frequency home appliances and includes the following steps.

[0257] S501. The sum of the torque compensation amount and the initial torque corresponding to the initial torque command is determined as the intermediate torque.

[0258] Receive torque compensation amount calculated from complex harmonic coefficients and compare it with the initial torque corresponding to the initial torque command. The results are summed to form an intermediate torque that reflects the superposition of compensation. .

[0259] S502. Based on the preset torque limit, the intermediate torque is limited to obtain the target torque corresponding to the target torque command.

[0260] Based on the preset torque limit, the intermediate torque is adjusted. Apply boundary constraints to prevent the compensated setpoint from exceeding the motor's allowable operating range, thereby obtaining the target torque that satisfies the control constraints. .

[0261] The preset torque limit is the preset upper limit of torque. and preset torque lower limit The preset torque limit is pre-set in the controller parameter area and loaded during power-on initialization, so that the limiting behavior has a fixed and reproducible numerical boundary during operation.

[0262] For intermediate torque The target torque is obtained by executing the preset torque limiting rule. To limit the output range of the torque channel, the preset torque limiting rule is implemented using upper and lower limit clamping.

[0263] When the intermediate torque Greater than or equal to the preset torque limit At that time, the preset torque upper limit will be set. Set as target torque When the intermediate torque Less than or equal to the preset lower torque limit At that time, the preset lower torque limit will be set. Set as target torque When the intermediate torque Less than the preset torque limit And greater than the preset lower limit of torque At that time, the intermediate torque Determined as target torque .

[0264] S503. Generate the target torque command based on the target torque.

[0265] The target torque is written into the torque command register or the given buffer, the target torque command is generated and output to the subsequent current control link, so that the motor maintains a stable torque command while maintaining the compensation effect.

[0266] In some embodiments, the torque conversion factor of the motor is determined; when the torque conversion factor is greater than or equal to the lower threshold, the target torque is divided by the torque conversion factor to obtain the target quadrature-axis current; when the torque conversion factor is less than the lower threshold, the target torque is divided by the lower threshold to obtain the target quadrature-axis current; and a target torque command corresponding to the target quadrature-axis current is generated.

[0267] Torque conversion coefficients can be calculated based on motor parameter tables or online identification results. And convert the torque to a coefficient. With lower threshold Compare.

[0268] When torque conversion factor Meets the lower threshold When required, the target quadrature-axis current is obtained by calculating the ratio of the target torque to the torque conversion factor. When the torque conversion factor Less than the lower threshold If so, then the lower threshold value should be used instead. Perform division to obtain the constrained target quadrature-axis current. .

[0269] The target quadrature-axis current can be written into the vector control link as the current setpoint, and a corresponding target torque command can be generated, so that the target torque command can directly participate in the subsequent current closed-loop regulation.

[0270] In this embodiment, the mapping relationship between the target torque and the target quadrature axis current can remain continuous under low coefficient conditions, reducing abnormal fluctuations in the current command caused by the torque conversion coefficient being too small, thereby improving the generation stability of the target torque command and providing a more consistent input basis for subsequent current control.

[0271] Specifically, the direct-axis inductance parameters, quadrature-axis inductance parameters, permanent magnet flux linkage parameters, number of pole pairs, and direct-axis current feedback of the motor are obtained; based on the difference between the direct-axis and quadrature-axis inductance parameters and the direct-axis current feedback, the inductance difference coupling term is determined; the sum of the inductance difference coupling term and the permanent magnet flux linkage parameter is determined as the equivalent flux linkage term; and the product of the preset coefficient, the number of pole pairs, and the equivalent flux linkage term is determined as the torque conversion coefficient.

[0272] Read the direct-axis inductance parameters in the motor parameter configuration area. Quadrature axis inductance parameters Permanent magnet flux linkage parameters and extreme pairs Simultaneously, the direct-axis current feedback quantity output by the current sampling circuit is acquired. .

[0273] Among them, direct-axis current feedback quantity This is the direct-axis current feedback quantity obtained from current sampling and coordinate transformation.

[0274] It can be based on quadrature axis inductance parameters With direct-axis inductance parameters The difference And combined with direct-axis current feedback Calculate the inductance difference coupling term Then, the inductance differential coupling term is combined with the permanent magnet flux linkage parameter. Adding them together yields the equivalent flux linkage term. .

[0275] Furthermore, the preset coefficients The number of pole pairs p and the equivalent magnetic flux terms Multiply to obtain the torque conversion factor. The preset coefficient can be set to a proportional constant corresponding to the torque constant of the three-phase motor, so that the conversion result is consistent with the unit system of the target control link.

[0276] Preset coefficients It can be In practical applications, other calibration values ​​can be selected for the preset coefficient, and this application does not limit this.

[0277] The target torque and target quadrature-axis current are converted based on the torque conversion factor, so that the torque command can be mapped to a current setpoint consistent with the current motor parameters. Since the equivalent flux linkage term takes into account the coupling effect of permanent magnet flux linkage and inductance difference, this conversion relationship can be updated synchronously with changes in motor parameters and used in the subsequent target torque command generation process.

[0278] In this embodiment, the torque conversion coefficient is determined jointly by real-time current feedback and motor parameters, ensuring that the conversion relationship between the target torque and current remains consistent with the actual operating state of the motor, thereby improving the consistency of target torque command generation and control accuracy. After the equivalent flux linkage term is used to characterize the inductive differential coupling, the conversion relationship more closely matches the actual electromagnetic characteristics of the salient-pole motor, which helps to improve the stability of torque control under different operating conditions.

[0279] The torque compensation method provided in this application, by first superimposing the torque compensation amount with the initial torque and then performing amplitude limiting processing, can keep the compensation result within an executable range and ensure consistency between periodic disturbance suppression and control safety boundaries for the target torque command. This method can improve the availability and closed-loop stability of compensation commands in variable frequency home appliance motor control scenarios, and reduce command distortion caused by excessive compensation.

[0280] Figure 6 This is a structural schematic diagram of an air conditioning control device according to some embodiments of this disclosure. Please refer to... Figure 6 The torque command compensation device 600 may include a data acquisition module 601, a mapping module 602, an analysis module 603, a determination module 604, and a compensation module 605.

[0281] The acquisition module 601 is used to acquire the operating characteristic parameters of the motor, including the initial torque command and the rotor electrical angle.

[0282] The mapping module 602 is used to perform complex domain mapping processing on the running feature parameters to obtain a composite input vector;

[0283] The analysis module 603 is used to perform feature analysis on the composite input vector through a complex neural network to generate a set of complex harmonic coefficients that correspond to the number of synthesis units at each order.

[0284] The determination module 604 is used to determine the torque compensation amount based on the set of complex harmonic coefficients and the rotor electrical angle.

[0285] The compensation module 605 is used to compensate the initial torque corresponding to the initial torque command by means of torque compensation amount, so as to obtain the compensated target torque command.

[0286] As an optional implementation, the operating characteristic parameters also include quadrature-axis current feedback; the analysis module 603 is specifically used for:

[0287] The estimated electromagnetic torque is determined based on the preset torque conversion factor and the quadrature axis current feedback.

[0288] Determine whether the initial torque command and estimated electromagnetic torque meet the enable conditions;

[0289] If so, the complex input vector is analyzed and processed through a complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

[0290] As an optional implementation, the analysis module 603 is specifically used for:

[0291] Determine the torque error between the indicated electromagnetic torque corresponding to the initial torque command and the estimated electromagnetic torque;

[0292] If the absolute value of the torque error is greater than or equal to the torque error threshold, then the enable condition is met.

[0293] If the absolute value of the torque error is less than the torque error threshold, the enable condition is not met.

[0294] As an optional implementation, the analysis module 603 is specifically used for:

[0295] If the absolute value of the torque error is greater than or equal to the torque error threshold, then update the number of consecutive cycles;

[0296] When the number of consecutive cycles is greater than or equal to the preset number of confirmation cycles, the enable condition is determined to be met.

[0297] As an optional implementation, the complex neural network includes multiple layers of fully connected complex layers, a discrete complex activation function, and a complex coefficient output header; the analysis module 603 is specifically used for:

[0298] A multi-layer complex fully connected layer is used to transform the composite input vector layer by layer to obtain the complex latent variable components.

[0299] By using a separate complex activation function, the components of the complex latent variable are processed to obtain the complex latent variable;

[0300] By using the complex coefficient output head, the complex latent variables are grouped and output in order to obtain a set of complex harmonic coefficients that corresponds to the number of synthesis units in order.

[0301] As an optional implementation, the operating characteristic parameters include torque command, rotor electrical angle, speed, direct-axis current feedback, and quadrature-axis current feedback; the mapping module 602 is specifically used for:

[0302] Determine the cosine component and sine component of the electrical angle corresponding to the rotor electrical angle;

[0303] The input feature vector is determined based on the torque command, speed, DC current feedback, quadrature axis current feedback, cosine component of electrical angle, and sine component of electrical angle.

[0304] Based on paired channels, the input feature vector is mapped to the complex field to obtain a composite input vector.

[0305] As an optional implementation, the complex harmonic coefficient set includes multiple complex harmonic coefficients, the number of which is equal to the number of successive synthesis units; the determining module 604 is specifically used for:

[0306] Based on the complex harmonic coefficients and rotor electrical angles corresponding to each preset order, determine the step-by-step compensation amount corresponding to each preset order.

[0307] The sum of the successive compensation amounts corresponding to each preset order is determined as the torque compensation amount.

[0308] As an optional implementation, the complex harmonic coefficients include the real part and the imaginary part of the complex harmonics; for any preset order; the determining module 604 is specifically used for:

[0309] Based on the preset order and rotor electrical angle, determine the order rotor cosine value and order rotor sine value;

[0310] Determine the first product of the real part of the complex harmonic and the order rotor cosine value;

[0311] Determine the second product of the imaginary part of the complex harmonic and the order rotor sine;

[0312] The difference between the first product and the second product is determined as the successive compensation amount.

[0313] As an optional implementation, the compensation module 605 is specifically used for:

[0314] The sum of the torque compensation amount and the initial torque corresponding to the initial torque command is determined as the intermediate torque;

[0315] Based on the preset torque limit, the intermediate torque is limited to obtain the target torque corresponding to the target torque command.

[0316] Generate a target torque command based on the target torque.

[0317] As an optional implementation, the compensation module 605 is specifically used for:

[0318] Determine the torque conversion factor for the motor;

[0319] When the torque conversion factor is greater than or equal to the lower threshold, the target torque is divided by the torque conversion factor to obtain the target quadrature axis current.

[0320] When the torque conversion factor is less than the lower threshold, the target torque is divided by the lower threshold to obtain the target quadrature axis current.

[0321] Generate the target torque command corresponding to the target quadrature axis current.

[0322] As an optional implementation, the compensation module 605 is specifically used for:

[0323] Obtain the direct-axis inductance parameters, quadrature-axis inductance parameters, permanent magnet flux linkage parameters, number of pole pairs, and direct-axis current feedback of the motor;

[0324] The inductance difference coupling term is determined based on the difference between the direct-axis inductance parameters and the quadrature-axis inductance parameters and the direct-axis current feedback.

[0325] The sum of the inductive differential coupling term and the permanent magnet flux linkage parameter is determined as the equivalent flux linkage term;

[0326] The product of the preset coefficients, the number of pole pairs, and the equivalent flux linkage term is determined as the torque conversion coefficient.

[0327] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0328] Figure 7 This is a block diagram illustrating an electronic device according to some embodiments of the present disclosure. For example, the electronic device 700 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0329] Reference Figure 7 The electronic device 700 may include one or more of the following components: processing component 702, memory 704, power component 707, multimedia component 708, audio component 710, input / output (I / O) interface 712, sensor component 714, and communication component 717.

[0330] Processing component 702 typically controls the overall operation of electronic device 700, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 702 may include one or more modules to facilitate interaction between processing component 702 and other components. For example, processing component 702 may include a multimedia module to facilitate interaction between multimedia component 708 and processing component 702.

[0331] Memory 704 is configured to store various types of data to support the operation of electronic device 700. Examples of this data include instructions for any application or method operating on electronic device 700, contact data, phonebook data, messages, pictures, videos, etc. Memory 704 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0332] Power component 706 provides power to various components of electronic device 700. Power component 706 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 700.

[0333] Multimedia component 708 includes a screen that provides an output interface between electronic device 700 and user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 708 includes a front-facing camera and / or a rear-facing camera. When electronic device 700 is in an operating mode, such as a shooting mode or video mode, the front-facing camera and / or rear-facing camera may receive multimedia data from outside. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0334] Audio component 710 is configured to output and / or input audio signals. For example, audio component 710 includes a microphone (MIC) configured to receive outdoor audio signals when electronic device 700 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 704 or transmitted via communication component 717. In some embodiments, audio component 710 also includes a speaker for outputting audio signals.

[0335] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0336] Sensor assembly 714 includes one or more sensors for providing state assessments of various aspects of electronic device 700. For example, sensor assembly 714 may detect the on / off state of electronic device 700, the relative positioning of components such as the display and keypad of electronic device 700, changes in position of electronic device 700 or a component of electronic device 700, the presence or absence of user contact with electronic device 700, orientation or acceleration / deceleration of electronic device 700, and temperature changes of electronic device 700. Sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 714 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0337] Communication component 716 is configured to facilitate wired or wireless communication between electronic device 700 and other devices. Electronic device 700 can access wireless networks based on communication standards, such as WiFi, 3G, 4G, 5G, other communication standards, or combinations thereof. In some embodiments of this disclosure, communication component 716 receives broadcast signals or broadcast-related information from an outdoor broadcast management system via a broadcast channel. In some embodiments of this disclosure, communication component 716 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0338] In some embodiments of this disclosure, the electronic device 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0339] In some embodiments of this disclosure, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, which can be executed by a processor 720 of an electronic device 700 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0340] A non-transitory computer-readable storage medium that, when instructions in the storage medium are executed by a processor of a mobile terminal, enables the mobile terminal to perform a torque compensation method.

[0341] Figure 8 This is a block diagram illustrating another electronic device according to some embodiments disclosed in a publication. For example, electronic device 800 can be provided as an air conditioner. (See also...) Figure 8 The electronic device 800 includes a processing component 822, which further includes one or more processors, and memory resources represented by a memory 832 for storing instructions, such as application programs, that can be executed by the processing component 822. The application programs stored in the memory 832 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 822 is configured to execute instructions to perform the torque compensation method described above.

[0342] The electronic device 800 may also include a power supply component 826 configured to perform power management of the device 800, a wired or wireless network interface 850 configured to connect the electronic device 800 to a network, and an input / output (I / O) interface 858. The device 800 may operate on an operating system stored in memory 832, such as Windows Server™, MacOS X™, Unix™, Linux™, FreeBSD™, or similar.

[0343] Some embodiments of this disclosure also provide a chip system, such as Figure 9 As shown, the chip system includes at least one processor 901 and at least one interface circuit 902. The processor 901 and the interface circuit 902 are interconnected via lines. For example, the interface circuit 902 can be used to receive signals from other devices (e.g., the memory of an electronic device). As another example, the interface circuit 902 can be used to send signals to other devices (e.g., the processor 901). Exemplarily, the interface circuit 902 can read instructions stored in memory and send those instructions to the processor 901. When the instructions are executed by the processor 901, the torque command compensation device can perform the steps in the above embodiments. Of course, the chip system may also include other discrete devices, and some embodiments of this disclosure do not specifically limit this.

[0344] In some embodiments of this disclosure, the interface circuit 902 can acquire data, program instructions, and / or information from the internal storage area of ​​the chip system; it can also acquire data, program instructions, and / or information from outside the chip system.

[0345] Optionally, the chip system also includes a memory 903, which is used to store necessary computer programs and data.

[0346] In some embodiments of this disclosure, an air conditioner is also provided for performing the torque compensation method provided in any of the above embodiments.

[0347] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this application can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this application.

[0348] In the above detailed description, reference has been made to the accompanying drawings, which illustrate specific aspects of this disclosure by way of illustration. In this regard, terms indicating direction or positional relationship, such as “center,” “longitudinal,” “lateral,” “length,” “width,” “thickness,” “upper,” “lower,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “top,” “bottom,” “inner,” “outer,” “clockwise,” “counterclockwise,” “axial,” “radial,” and “circumferential,” are used with reference to the orientation of the described figures. Since components of the described device can be positioned in multiple different orientations, directional terms are used for illustrative purposes and not for limitation. It should be understood that other aspects can be utilized and structural or logical changes can be made without departing from the concept of this disclosure. Therefore, the following detailed description should not be considered limiting.

[0349] It should be understood that, unless otherwise specifically indicated, features of various embodiments of this disclosure described herein can be combined with each other. As used herein, the term “and / or” includes any one of the relevant listed items and any combination of any two or more; similarly, “at least one of…” includes any one of the relevant listed items and any combination of any two or more.

[0350] It should be understood that, unless otherwise expressly specified and limited, the terms "joining," "attaching," "installing," "connecting," "linking," "fixing," etc., used in the embodiments of this disclosure should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms herein based on the specific circumstances.

[0351] Furthermore, the term "above" as used herein with respect to components, elements, or material layers formed or located "above" a surface may be used to indicate that the component, element, or material layer is "indirectly" positioned (e.g., placed, formed, deposited, etc.) on the surface such that one or more additional components, elements, or layers are arranged between the surface and the component, element, or material layer. However, the term "above" as used with respect to components, elements, or material layers formed or located "above" a surface may also optionally have a specific meaning: that the component, element, or material layer is "directly" positioned (e.g., placed, formed, deposited, etc.) on the surface, for example, in direct contact with the surface.

[0352] Although terms such as “first,” “second,” and “third” may be used herein to describe various components, parts, regions, layers, or sections, these components, parts, regions, layers, or sections are not limited to these terms. Rather, these terms are used only to distinguish one component, part, region, layer, or section from another. Therefore, without departing from the teachings of the examples described herein, the first component, part, region, layer, or section mentioned in the examples may also be referred to as the second component, part, region, layer, or section. Furthermore, the terms “first” and “second” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as “first” or “second” may explicitly or implicitly include at least one of that feature. In the description herein, “a plurality” means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0353] It should be understood that spatial relative terms, such as “above,” “upper,” “below,” and “lower,” are used herein to describe the relationship between one element and another shown in the figures. In addition to the orientation depicted in the figures, these spatial relative terms are also intended to encompass different orientations of the device in use or operation. For example, if the device in the figures is flipped, an element described as “above” or “upper” relative to another element would be “below” or “lower” relative to that other element. Thus, depending on the spatial orientation of the device, the term “above” encompasses both above and below orientations. Devices may have other orientations (e.g., rotated 90 degrees or in other orientations), and the spatial relative terms used herein should be interpreted accordingly.

[0354] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”

[0355] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”

[0356] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0357] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A torque compensation method, characterized in that, include: Collect the motor's operating characteristic parameters, including the initial torque command and rotor electrical angle; The operational feature parameters are subjected to complex domain mapping to obtain a composite input vector; The composite input vector is processed by a complex neural network to generate a set of complex harmonic coefficients that correspond to the number of synthesis units at each order. The torque compensation amount is determined based on the set of complex harmonic coefficients and the rotor electrical angle. The initial torque corresponding to the initial torque command is compensated by the torque compensation amount to obtain the compensated target torque command.

2. The method according to claim 1, characterized in that, The operating characteristic parameters also include the quadrature-axis current feedback; the composite input vector is processed by a complex neural network to generate a set of complex harmonic coefficients corresponding to the number of successive synthesis units, including: The estimated electromagnetic torque is determined based on the preset torque conversion factor and the cross-axis current feedback. Determine whether the initial torque command and the estimated electromagnetic torque satisfy the enable condition; If so, the complex input vector is analyzed and processed through the complex neural network to generate a set of complex harmonic coefficients that corresponds to the number of synthesis units at each order.

3. The method according to claim 2, characterized in that, Determining whether the initial torque command and the estimated electromagnetic torque satisfy the enable condition includes: Determine the torque error between the indicated electromagnetic torque corresponding to the initial torque command and the estimated electromagnetic torque; If the absolute value of the torque error is greater than or equal to the torque error threshold, then the enabling condition is satisfied; If the absolute value of the torque error is less than the torque error threshold, then the enabling condition is not met.

4. The method according to claim 3, characterized in that, If the absolute value of the torque error is greater than or equal to the torque error threshold, then the enabling condition is satisfied, including: If the absolute value of the torque error is greater than or equal to the torque error threshold, then update the number of consecutive cycles; When the number of consecutive cycles is greater than or equal to the preset number of confirmation cycles, the enabling condition is determined to be met.

5. The method according to claim 2, characterized in that, The complex neural network includes multiple layers of fully connected complex layers, a discrete complex activation function, and a complex coefficient output header. Through the complex neural network, the composite input vector is analyzed and processed to generate a set of complex harmonic coefficients corresponding to the order of the number of synthesis units, including: The composite input vector is transformed layer by layer using the multi-layer complex fully connected layer to obtain complex latent variable components. The complex latent variable components are processed by the separated complex activation function to obtain the complex latent variables; By using the complex coefficient output head, the complex latent variables are grouped and output in order to obtain a set of complex harmonic coefficients that corresponds to the number of the order synthesis units in order.

6. The method according to any one of claims 1-5, characterized in that, The operating characteristic parameters include torque command, rotor electrical angle, speed, direct-axis current feedback, and quadrature-axis current feedback. The operational feature parameters are subjected to complex domain mapping to obtain a composite input vector, including: Determine the cosine component and sine component of the electrical angle corresponding to the rotor electrical angle; The input feature vector is determined based on the torque command, the rotational speed, the DC current feedback, the quadrature axis current feedback, the cosine component of the electrical angle, and the sine component of the electrical angle. Based on paired channels, the input feature vector is mapped to the complex field to obtain the composite input vector.

7. The method according to any one of claims 1-5, characterized in that, The set of complex harmonic coefficients includes multiple complex harmonic coefficients, the number of which is equal to the number of successive synthesis units; the torque compensation amount is determined based on the set of complex harmonic coefficients and the rotor electrical angle, including: The step-by-step compensation amount corresponding to each preset order is determined based on the complex harmonic coefficients corresponding to each preset order and the rotor electrical angle. The sum of the successive compensation amounts corresponding to each preset order is determined as the torque compensation amount.

8. The method according to claim 7, characterized in that, The complex harmonic coefficients include the real part and the imaginary part of the complex harmonics; for any preset order; based on the complex harmonic coefficients corresponding to the preset order and the rotor electrical angle, the successive compensation amount corresponding to the preset order is determined, including: Based on the preset order and the rotor electrical angle, determine the order rotor cosine value and the order rotor sine value; Determine the first product of the real part of the complex harmonic and the order rotor cosine value; Determine the second product of the imaginary part of the complex harmonic and the order rotor sine value; The difference between the first product and the second product is determined as the successive compensation amount.

9. The method according to any one of claims 1-5, characterized in that, The initial torque corresponding to the initial torque command is compensated using the torque compensation amount to obtain the compensated target torque command, including: The sum of the torque compensation amount and the initial torque corresponding to the initial torque command is determined as the intermediate torque; Based on a preset torque limit, the intermediate torque is limited to obtain the target torque corresponding to the target torque command; The target torque command is generated based on the target torque.

10. The method according to claim 9, characterized in that, Based on the target torque, the target torque command is generated, including: Determine the torque conversion factor of the motor; When the torque conversion factor is greater than or equal to the lower threshold, the target torque is divided by the torque conversion factor to obtain the target quadrature axis current; When the torque conversion factor is less than the lower threshold, the target torque is divided by the lower threshold to obtain the target quadrature axis current; Generate the target torque command corresponding to the target quadrature axis current.

11. The method according to claim 10, characterized in that, Determining the torque conversion factor of the motor includes: Obtain the direct-axis inductance parameters, quadrature-axis inductance parameters, permanent magnet flux linkage parameters, number of pole pairs, and direct-axis current feedback of the motor; The inductance difference coupling term is determined based on the difference between the direct-axis inductance parameters and the quadrature-axis inductance parameters and the direct-axis current feedback. The sum of the inductance differential coupling term and the permanent magnet flux linkage parameter is determined as the equivalent flux linkage term; The product of the preset coefficient, the number of pole pairs, and the equivalent flux linkage term is determined as the torque conversion coefficient.

12. An air conditioner, characterized in that, Used to perform the method as described in any one of claims 1-11.

13. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to execute the executable instructions to implement the method of any one of claims 1-11.

14. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor, the processor is able to perform the method as described in any one of claims 1-11.

15. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-11.

16. A chip system, characterized in that, The chip system includes a processing unit and an interface circuit. The processing unit obtains program instructions through the interface circuit, and the program instructions are executed by the processing unit. The processing unit is used to perform the method as described in any one of claims 1-11.

Images