Motor control method and apparatus, and top-discharge outdoor unit

Abstract

The present application discloses a motor control method, the method comprising: acquiring an input current data set and a rotation data set of a motor of a top-discharge outdoor unit, wherein the motor is configured to drive a fan blade of the top-discharge outdoor unit to rotate, the input current data set comprises a plurality of pieces of input current data, and the rotation data set comprises a plurality of pieces of rotation data; inputting the input current data set and the rotation data set into a fan blade state determination model, and processing the input current data set and the rotation data set via the fan blade state determination model to obtain a fan blade state of the fan blade, the fan blade state comprising balanced and unbalanced; and controlling the motor on the basis of the fan blade state of the fan blade.

Description

Control method and device of motor and ejector outdoor unit

[0001] The present application claims priority to the Chinese patent application No. 2024113492031, filed on September 25, 2024, and entitled "Control method and device of motor and ejector outdoor unit", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of motor control, and in particular to a control method and device of motor and an ejector outdoor unit. BACKGROUND

[0003] For the ejector outdoor unit, under the influence of extreme weather such as severe cold and freezing rain, the fan blades of the ejector outdoor unit may freeze ice on the surface due to water vapor condensation and other factors. If the distribution of ice on the fan blades is uneven at this time, the fan blades may be unbalanced due to uneven distribution of the mass of the fan blades after the motor starts, which may cause problems such as unstable operation of the motor, overcurrent shutdown, and even damage to the fan blades during high-speed operation of the motor.

[0004] Therefore, how to identify the fan blade state of the fan blades and control the motor according to the fan blade state is a research hotspot. SUMMARY

[0005] Embodiments of the present application provide a control method and device of motor and an ejector outdoor unit, which can identify the fan blade state of the fan blades and control the motor according to the fan blade state. The technical solutions are as follows:

[0006] In one aspect, a control method of motor is provided, the method comprising:

[0007] obtaining an input current data set and a rotation data set of a motor of an ejector outdoor unit, the motor being configured to drive fan blades of the ejector outdoor unit to rotate, the input current data set comprising a plurality of input current data, and the rotation data set comprising a plurality of rotation data;

[0008] inputting the input current data set and the rotation data set into a fan blade state determination model, processing the input current data set and the rotation data set by the fan blade state determination model to obtain a fan blade state of the fan blades, the fan blade state comprising balance and imbalance;

[0009] controlling the motor based on the fan blade state of the fan blades.

[0010] In one aspect, a control device of motor is provided, the device comprising:

[0011] a data acquisition module configured to acquire a set of input current data of a motor of the ejecting air outdoor unit and a set of rotation data, the motor being configured to drive a fan blade of the ejecting air outdoor unit to rotate, the set of input current data comprising a plurality of input current data, the set of rotation data comprising a plurality of rotation data;

[0012] a fan blade state determination module configured to input the set of input current data and the set of rotation data into a fan blade state determination model, process the set of input current data and the set of rotation data by the fan blade state determination model, and obtain a fan blade state of the fan blade, the fan blade state comprising balance and imbalance;

[0013] a control module configured to control the motor based on the fan blade state of the fan blade.

[0014] In a possible implementation, the rotation data comprises rotation speed and rotation position of a rotor of the motor, the fan blade state determination module is configured to transform corresponding input current data in the set of input current data based on a plurality of rotation positions in the set of rotation data by the fan blade state determination model, to obtain a plurality of first current pairs of the motor, one of the first current pairs comprising a set of d-axis current and q-axis current; and determine the fan blade state of the fan blade based on the plurality of first current pairs of the motor and a plurality of rotation speeds in the set of rotation data by the fan blade state determination model.

[0015] In a possible implementation, the fan blade state determination module is configured to perform Clarke transformation on each input current data in the set of input current data by the fan blade state determination model, to obtain a plurality of second current pairs, one of the second current pairs comprising a set of α current and β current in a two-phase static coordinate system; and perform Park transformation on corresponding second current pair in the plurality of second current pairs based on a plurality of rotation positions in the set of rotation data by the fan blade state determination model, to obtain the plurality of first current pairs of the motor.

[0016] In a possible implementation, the fan blade state determination module is configured to perform feature extraction on the plurality of first current pairs and the plurality of rotation speeds in the set of rotation data by the fan blade state determination model, to obtain balance feature of the fan blade; perform full connection and normalization on the balance feature by the fan blade state determination model, to obtain a fan blade state prediction score of the fan blade; in a case where the fan blade state prediction score is greater than or equal to a state score threshold, determine the fan blade state of the fan blade as balance; and in a case where the fan blade state prediction score is less than the state score threshold, determine the fan blade state of the fan blade as imbalance.

[0017] In a possible implementation, the blade state determination module is configured to normalize and perform time-frequency transformation on the plurality of first current pairs and the plurality of rotation speeds in the rotation data set to obtain current spectra of the plurality of first current pairs and rotation speed spectra of the plurality of rotation speeds by using the blade state determination model; perform feature extraction on the current spectra and the rotation speed spectra to obtain first spectral features of the rotation speed spectra and second spectral features of the rotation speed spectra by using the blade state determination model; and fuse the first spectral features and the second spectral features to obtain the balance feature of the blade by using the blade state determination model.

[0018] In a possible implementation, the blade state determination module is configured to determine a first fusion weight corresponding to the first spectral features and a second fusion weight corresponding to the second spectral features by using the blade state determination model, the first fusion weight being configured to represent a degree of association between a first current pair and a blade state of the blade, and the second fusion weight being configured to represent a degree of association between a rotation speed and the blade state of the blade; and fuse the first spectral features and the second spectral features by using the first fusion weight and the second fusion weight to obtain the balance feature of the blade.

[0019] In a possible implementation, the blade state determination module is configured to perform time sequence encoding on the plurality of first current pairs and the plurality of rotation speeds to obtain current time sequence features of the plurality of first currents and rotation speed time sequence features of the plurality of rotation speeds by using the blade state determination model; and fuse the current time sequence features and the rotation speed time sequence features to obtain the balance feature of the blade by using the blade state determination model.

[0020] In a possible implementation, the control module is configured to control the motor to continue rotating according to a current control parameter in a case where the blade state of the blade is balanced, and control the motor to reduce a rotating speed or stop rotating in a case where the blade state of the blade is unbalanced.

[0021] In a possible implementation, the control module is configured to determine a number of times that the blade state of the blade is continuously determined to be unbalanced in a case where the blade state of the blade is unbalanced, control the motor to reduce the rotating speed in a case where the number is greater than or equal to a first preset number, and control the motor to stop rotating in a case where the number is greater than or equal to a second preset number, the second preset number being greater than the first preset number.

[0022] In a possible implementation, the device further includes a training module configured to obtain a plurality of sample data sets of the top-out outdoor unit and corresponding labeled blade states of each of the sample data sets, the sample data sets including a sample input current data set and a sample rotation data set; input the plurality of sample data sets into the blade state determination model, process each of the sample data sets by the blade state determination model to obtain a predicted blade state corresponding to each of the sample data sets; and train the blade state determination model based on difference information between the predicted blade state and the labeled blade state corresponding to each of the sample data sets.

[0023] In a possible implementation, the training module is configured to obtain a plurality of first sample data sets of the top-out outdoor unit, the plurality of first sample data sets being collected in a case where the blade state of the blade is balanced, different first sample data sets being collected in a case where the top-out outdoor unit is in different backwind blocking ratios; and obtain a plurality of second sample data sets of the top-out outdoor unit, the plurality of second sample data sets being collected in a case where the blade state of the blade is unbalanced, different second sample data sets being collected in a case where the top-out outdoor unit is in different unbalanced loads.

[0024] In an aspect, a top-out outdoor unit is provided, which includes one or more processors and one or more memories, the one or more memories storing at least one computer program, the computer program being loaded and executed by the one or more processors to implement the control method of the motor.

[0025] In an aspect, a computer readable storage medium is provided, which stores at least one computer program, the computer program being loaded and executed by a processor to implement the control method of the motor.

[0026] In an aspect, a computer program product or computer program is provided, which includes program code stored in a computer readable storage medium, a processor of a computer device reading the program code from the computer readable storage medium, and the processor executing the program code to cause the computer device to perform the control method of the motor. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0028] Fig. 1 is a structural schematic diagram of a top-out air chamber outdoor unit according to an embodiment of the present application;

[0029] Fig. 2 is a flow chart of a motor control method according to an embodiment of the present application;

[0030] Fig. 3 is a flow chart of another motor control method according to an embodiment of the present application;

[0031] Fig. 4 is a schematic diagram of data comparison before and after normalization according to an embodiment of the present application;

[0032] Fig. 5 is a flow chart of yet another motor control method according to an embodiment of the present application;

[0033] Fig. 6 is a flow chart of a wind blade state determination model training method according to an embodiment of the present application;

[0034] Fig. 7 is a schematic diagram of a hyperplane according to an embodiment of the present application;

[0035] Fig. 8 is a processing result schematic diagram of a kernel function according to an embodiment of the present application;

[0036] Fig. 9 is a processing result schematic diagram of a Gaussian kernel function according to an embodiment of the present application;

[0037] Fig. 10 is a structural schematic diagram of a motor control device according to an embodiment of the present application;

[0038] Fig. 11 is a structural schematic diagram of another top-out air chamber outdoor unit according to an embodiment of the present application. DETAILED DESCRIPTION

[0039] To make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in further detail below with reference to the drawings.

[0040] In the present application, the terms "first", "second", etc. are used to distinguish the same or similar items with basically the same function, and it should be understood that there is no logical or time sequence dependency between "first", "second", "nth", and the number and execution order are not limited.

[0041] To describe the technical solutions provided by the embodiments of the present application, some terms related to the embodiments of the present application will be introduced below.

[0042] Top-out air-cooled outdoor unit: Top-out air-cooled outdoor unit is a part of central air conditioning system, mainly used for large-scale building cooling and heating needs. This type of outdoor unit has its air outlet facing upwards, usually installed on the roof or balcony of the building. The main function of the top-out air-cooled outdoor unit is to exchange heat between indoor and outdoor through refrigerant circulation, providing a suitable temperature environment for indoor.

[0043] Fan blade: The main function of the fan blade is to accelerate air flow, helping the outdoor unit to exchange heat. In cooling mode, the fan is used to accelerate air through the condenser, taking away the heat in it, so that the refrigerant condenses into a liquid; while in heating mode, it helps to increase air flow, assisting the condenser to release heat.

[0044] D-axis current / q-axis current: In permanent magnet synchronous motor, d-axis and q-axis are virtual coordinate systems established on the basis of rotor magnetic field. The d-axis current is consistent with the direction of the rotor magnetic field, while the q-axis current is perpendicular to the direction of the rotor magnetic field. The currents of these two axes are used to control the magnetic field and torque of the motor, achieving precise control of the motor performance.

[0045] Clark transformation: Also known as Clark's transformation, it is a very important technology in motor control, mainly used to convert three-phase static coordinate system current or voltage into equivalent quantities in two-phase static coordinate system.

[0046] Park transformation: It is a commonly used coordinate transformation method in motor control, which is used to convert the time domain components of three-phase current or voltage into two components in the orthogonal stationary coordinate system, namely d-axis and q-axis components.

[0047] Machine Learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is a discipline that studies how computers simulate or implement human learning behavior to acquire new knowledge or skills, reorganize existing knowledge sub-models to continuously improve their performance.

[0048] Support Vector Machine (SVM): It is a supervised learning binary classification model, whose basic model is a linear classifier with maximum margin in feature space. In two-dimensional space, the hyperplane can be a straight line; in three-dimensional space, it is a plane; and in higher-dimensional space, it is a hyperplane. The core idea of SVM is to find a hyperplane that maximizes the margin between the two classes of samples. When the data set is linearly separable, SVM can find an optimal hyperplane; when the data set is approximately linearly separable or non-linearly separable, SVM uses kernel functions to map the data to a higher-dimensional space, making the data linearly separable in that space, thereby achieving classification.

[0049] Fourier transform: Fourier transform is an important mathematical tool that can convert signals from time domain to frequency domain, so that we can analyze the frequency characteristics of the signal.

[0050] Normalization: The value range of different number series is mapped to (0, 1) or (-1, 1) interval, which is convenient for data processing. In some cases, the normalized numerical value can be directly implemented as a probability.

[0051] The control method for the motor of the top-out air-cooled unit in the related art generally cannot identify whether the fan blade is balanced, which is easy to cause damage to the fan blade or the motor during the operation of the top-out air-cooled unit. After the technical solution provided in the embodiments of the present application is adopted, the fan blade state of the fan blade can be identified, so that the motor of the top-out air-cooled unit is controlled according to the fan blade state of the fan blade, thereby reducing the probability of damage to the fan blade and the motor due to unbalanced fan blades.

[0052] FIG. 1 is a schematic diagram of a top-out air-cooled unit provided in an embodiment of the present application, referring to FIG. 1, the top-out air-cooled unit includes a side wall panel 1, an air inlet 2, and a fan blade 3. The side wall panel 1 serves to support and protect, the air inlet 2 is used to suck air into the inside of the top-out air-cooled unit, and the fan blade 3 is used to accelerate air flow and help the top-out air-cooled unit to exchange heat. Since the fan blade 3 of the top-out air-cooled unit is vertically upward, under the influence of extreme weather such as severe cold and freezing rain, the fan blade 3 of the top-out air-cooled unit may be iced on the surface due to water vapor condensation and other factors. If the distribution of ice on the fan blade is uneven, the fan blade will be unbalanced.

[0053] FIG. 2 is a flowchart of a control method of a motor provided in an embodiment of the present application, referring to FIG. 2, taking the motor controller of the top-out air-cooled unit as an example, the method includes the following steps.

[0054] 201, the motor controller acquires an input current data set and a rotation data set of the motor of the top-out air-cooled unit, the motor being used to drive the fan blade of the top-out air-cooled unit to rotate, the input current data set including a plurality of input current data, and the rotation data set including a plurality of rotation data.

[0055] The motor is connected to the fan blade, which rotates under the motor's drive. The connection between the motor and the fan blade can be direct or indirect. Direct connection refers to the motor shaft being directly connected to the fan blade shaft. In some embodiments, flange or bevel gear connections are used. Direct connection offers advantages such as simple structure, compactness, and high reliability. Indirect connection involves connecting the motor and fan blade via a coupling or other components. Indirect connection reduces and balances vibration and impact between the motor and fan blade through the elastic buffer of the coupling, improving stability. This application does not limit the connection method between the motor and the fan blade. Input current data refers to the relevant data of the three-phase electricity directly input to the motor. In some embodiments, the input current data includes the current of each of the three phases. Rotation data refers to relevant data describing the motor's rotation. In some embodiments, active data includes the motor's rotational speed and the rotor's rotational position. The input current data set is a collection of multiple input current data, which are collected within a preset time period during the operation of the top-discharge outdoor unit. Similarly, the rotation data set is a collection of multiple rotation data, which are collected within a preset time period during the operation of the top-discharge outdoor unit. The preset time period is set by technicians according to the actual situation, and this application embodiment does not limit it.

[0056] 202. The motor controller inputs the input current data set and the rotation data set into the fan blade state determination model. The fan blade state determination model processes the input current data set and the rotation data set to obtain the fan blade state, which includes balanced and unbalanced states.

[0057] The blade state determination model is a binary classification model. It is trained based on multiple sample datasets of the top-discharge outdoor unit and the corresponding labeled blade states for each dataset. The sample datasets include sample input current data sets and sample rotation data sets. After training, the model learns the relationship between the input current data sets, rotation data sets, and the blade state of the unit. Therefore, it can classify the blades as balanced or unbalanced based on the multiple input current and rotation data.

[0058] 203. The motor controller controls the motor based on the state of the fan blade.

[0059] In this context, control refers to controlling the rotation of the motor. In some embodiments, controlling the motor to reduce its speed, stop rotating, or continue rotating according to the current control parameters can all be referred to as control. The current control parameters refer to the parameters used to control the motor determined under the conventional control logic.

[0060] By the technical solution provided in the embodiments of the present application, multiple input current data and multiple rotation data of the motor driving the top-out air chamber outdoor unit fan blade are acquired. The multiple input current data and multiple rotation data are processed by using a fan blade state determination model to determine the fan blade state of the fan blade, so as to realize the identification of the fan blade state of the fan blade. The motor of the top-out air chamber outdoor unit is controlled based on the fan blade state, thereby reducing the probability of damage of the fan blade and the motor due to fan blade imbalance.

[0061] The above steps 201-203 are a simple description of the motor control method provided in the embodiments of the present application. In the following, the motor control method provided in the embodiments of the present application will be described in more detail in combination with some examples. Referring to FIG. 3, taking the motor controller of the top-out air chamber outdoor unit as an example, the method includes the following steps.

[0062] 301. The motor controller acquires an input current data set and a rotation data set of the motor of the top-out air chamber outdoor unit, the motor being used to drive the fan blade of the top-out air chamber outdoor unit to rotate, the input current data set including multiple input current data, and the rotation data set including multiple rotation data.

[0063] The motor is connected with the fan blade, and the fan blade rotates under the driving of the motor. The connection between the motor and the fan blade includes direct connection and indirect connection. Direct connection means that the motor shaft of the motor is directly connected with the fan blade shaft of the fan blade. In some embodiments, a flange connection or a bevel gear connection is used. Direct connection has the advantages of simple structure, compactness and high reliability. Indirect connection means that the motor is connected with the fan blade through a coupling or the like. Indirect connection can reduce and balance the vibration and impact between the motor and the fan blade through the elastic buffering of the coupling, thereby improving stability. The embodiments of the present application do not limit the connection mode between the motor and the fan blade. The top-out air chamber outdoor unit is powered by three-phase electricity. Input current data refers to the relevant data of the three-phase electricity directly input to the motor. In some embodiments, the input current data includes the currents of U phase, V phase and W phase in the three-phase electricity. Rotation data refers to the relevant data describing the rotation of the motor. In some embodiments, the active data includes the rotation speed of the motor and the rotation position of the rotor.

[0064] In some embodiments, the motor controller acquires multiple input current data and multiple rotation data of the motor within a preset time length before the current time. The motor controller packs the multiple input current data into the input current data set and packs the multiple rotation data into the rotation data set.

[0065] In this implementation, the preset time length is a time length before the current time, and in some embodiments, the preset time length is 30s, 60s, or 180s, etc. before the current time, which is not limited in the embodiments of the present application. Packing the plurality of input current data into the input current data set means packing the plurality of input current data into an input current data sequence in the order of collection time, and the input current data sequence can reflect the change of the input current data of the motor over time. Similarly, packing the plurality of rotation data into the rotation data set means packing the plurality of rotation data into a rotation data sequence in the order of collection time, and the rotation data sequence can reflect the change of the rotation data of the motor over time. In some embodiments, the number of input current data in the input current data set is the same as the number of rotation data in the rotation data set, and any input current data in the input current data set has corresponding rotation data in the rotation data set, where the corresponding means the same or close collection time.

[0066] In this implementation, the plurality of input current data and the plurality of rotation data of the motor acquired within the preset time length before the current time are packed into the input current data set and the rotation data set, which can not only reflect the change of the data history, but also eliminate the error caused by a single data, and is helpful for subsequent identification of the fan blade state using the input current data and the rotation data.

[0067] In some embodiments, the input current data includes the current of each of the three phases of the three-phase power, and the rotation data includes the rotation speed and the rotation position (or rotation angle) of the rotor. The motor controller queries the storage medium using the preset time length to obtain the plurality of input current data and the plurality of rotation data within the preset time length before the current time. The motor controller packs the plurality of input current data into the input current data set in the order of collection time, and packs the plurality of rotation data into the rotation data set in the order of collection time.

[0068] It should be noted that since the sum of the currents of the three phases of the three-phase power is 0, actually, acquiring the current of any two phases of the three-phase power is equivalent to acquiring the current of each of the three phases. In some embodiments, the motor controller can acquire the current of the U phase and the current of the V phase to calculate the current of the W phase.

[0069] Another implementation of the above step 301 will be described below.

[0070] In some embodiments, the motor controller acquires a plurality of input current data and a plurality of rotation data of the motor within a preset time length after the current time. The motor controller packs the plurality of input current data into an input current data set and packs the plurality of rotation data into a rotation data set.

[0071] In this implementation, the preset duration is a period of time after the current moment. In some embodiments, the preset duration is 30 seconds, 60 seconds, or 180 seconds after the current moment, etc. This application does not limit this.

[0072] In this implementation, multiple input current data and multiple rotation data of the motor within a preset time period after the current moment are packaged into an input current data set and a rotation data set. This not only reflects the latest changes in the data, but also eliminates the errors caused by individual data, which helps to identify the state of the fan blades by using the input current data and rotation data in the future.

[0073] In some embodiments, the input current data includes the current of each of the three phases of the three-phase electricity, and the rotation data includes the rotation speed and the rotation position of the rotor. The motor controller continuously acquires the input current data and rotation data over a preset time period starting from the current moment, obtaining multiple input current data and multiple rotation data over the preset time period after the current moment. The motor controller packages the multiple input current data into an input current data set according to the order of acquisition time, and packages the multiple rotation data into a rotation data set according to the order of acquisition time.

[0074] Optionally, before obtaining the input current data set and rotation data set through the above implementation method, the motor controller can also preprocess the acquired multiple input current data and multiple rotation data to eliminate erroneous data and data spikes in the multiple input current data and multiple rotation data, thereby improving the accuracy of subsequent determination of the fan blade state.

[0075] 302. The motor controller inputs the input current data set and the rotation data set into the fan blade state determination model. The fan blade state determination model processes the input current data set and the rotation data set to obtain the fan blade state, which includes balanced and unbalanced states.

[0076] The blade state determination model is a binary classification model. It is trained based on multiple sample data sets of the top-discharge outdoor unit and the corresponding labeled blade states for each data set. The sample data sets include sample input current data sets and sample rotation data sets. After training, the model learns the relationship between the input current data sets, rotation data sets, and the blade states of the unit. Therefore, it can classify the blades as balanced or unbalanced based on the multiple input current and rotation data. The training process of this blade state determination model will be described in subsequent embodiments.

[0077] In some embodiments, the rotation data includes rotation speeds and rotation positions of a rotor of the motor, the motor controller inputs the input current data set and the rotation data set into the blade state determination model, through the blade state determination model, corresponding input current data in the input current data set is transformed based on multiple rotation positions in the rotation data set, to obtain multiple first current pairs of the motor, one first current pair includes a group of d-axis currents and q-axis currents. The motor controller determines the blade state of the blade through the blade state determination model based on the multiple first current pairs of the motor and the multiple rotation speeds in the rotation data set.

[0078] In some embodiments, the rotation data includes rotation speeds and rotation positions of a rotor of the motor, the motor controller inputs the input current data set and the rotation data set into the blade state determination model, through the blade state determination model, corresponding input current data in the input current data set is transformed based on multiple rotation positions in the rotation data set, to obtain multiple first current pairs of the motor, one first current pair includes a group of d-axis currents and q-axis currents. The motor controller determines the blade state of the blade through the blade state determination model based on the multiple first current pairs of the motor and the multiple rotation speeds in the rotation data set.

[0079] In some embodiments, the rotation data includes rotation speeds and rotation positions of a rotor of the motor, the motor controller inputs the input current data set and the rotation data set into the blade state determination model, through the blade state determination model, corresponding input current data in the input current data set is transformed based on multiple rotation positions in the rotation data set, to obtain multiple first current pairs of the motor, one first current pair includes a group of d-axis currents and q-axis currents. The motor controller determines the blade state of the blade through the blade state determination model based on the multiple first current pairs of the motor and the multiple rotation speeds in the rotation data set.

[0080] In order to make the above-mentioned embodiments more clearly, the following will be divided into several parts to explain the above-mentioned embodiments.

[0081] The first part is to explain the way of obtaining the multiple first current pairs of the motor through the blade state determination model.

[0082] In some embodiments, the motor controller performs Clarke transformation on each input current data in the input current data set through the blade state determination model to obtain multiple second current pairs, one second current pair includes a group of α currents and β currents in a two-phase static coordinate system. The motor controller performs Park transformation on corresponding second current pairs in the multiple second current pairs based on multiple rotation positions in the rotation data set through the blade state determination model to obtain the multiple first current pairs of the motor.

[0083] The Clark transformation and the Park transformation are both coordinate system transformation manners, and the three-phase current can be converted into the d-axis current and the q-axis current through the Clark transformation and the Park transformation.

[0084] In some embodiments, the input current data includes the current of the U phase, the current of the V phase, and the current of the W phase. For any input current data in the input current data set, the motor controller converts the current of the U phase, the current of the V phase, and the current of the W phase into the alpha current and the beta current through the fan blade state determination model, to obtain a first second current pair, by using formula (1) which is a transformation formula corresponding to the Clark transformation. The motor controller converts the alpha current and the beta current into the d-axis current and the q-axis current through the fan blade state determination model, to obtain a first current pair, by using formula (2).

[0085] wherein I U is the U-phase current, I V is the V-phase current, I W is the W-phase current, I α is the alpha current, I β is the beta current, I q is the q-axis current, I d is the d-axis current, and θ is the rotation position.

[0086] The second part describes the manner of determining the fan blade state of the fan blade.

[0087] In some embodiments, the motor controller extracts features from the plurality of first current pairs and the plurality of rotation speeds in the rotation data set through the fan blade state determination model to obtain the balance feature of the fan blade. The motor controller performs full connection and normalization on the balance feature through the fan blade state determination model to obtain the fan blade state prediction score of the fan blade. In a case where the fan blade state prediction score is greater than or equal to a state score threshold, the motor controller determines the fan blade state of the fan blade as balanced. In a case where the fan blade state prediction score is less than the state score threshold, the motor controller determines the fan blade state of the fan blade as unbalanced.

[0088] The balance feature is a high-dimensional expression of the plurality of first current pairs and the plurality of rotation speeds, and the balance feature can reflect the fan blade state of the fan blade from a higher dimension compared to the plurality of first current pairs and the plurality of rotation speeds. The state score threshold is set by a technician according to actual conditions, and embodiments of the present application do not limit this.

[0089] In order to more clearly illustrate the above embodiments, the manner of feature extraction in the above embodiments will be further described below.

[0090] In some embodiments, the motor controller normalizes and performs time-frequency transformation on the plurality of first current pairs and the plurality of rotating speeds in the rotating data set to obtain current spectra of the plurality of first current pairs and rotating speed spectra of the plurality of rotating speeds through the fan blade state determination model. The motor controller extracts features from the current spectra and the rotating speed spectra to obtain first spectral features of the rotating speed spectra and second spectral features of the rotating speed spectra through the fan blade state determination model. The motor controller fuses the first spectral features and the second spectral features to obtain the balance feature of the fan blade through the fan blade state determination model.

[0091] In some embodiments, the normalized data are all within the range of (-1, 1), as shown in FIG. 4. Before normalization, the current IU of the U phase, the current IV of the V phase, the rotating speed, the d-axis current Id, the q-axis current Iq, and the rotating position theta all have different value ranges. After normalization, they all fall within the interval of (-1, 1), which is convenient for subsequent processing. The time-frequency transformation is to convert the plurality of first current pairs and the plurality of rotating speeds in the time domain to the frequency domain, so as to reduce the difficulty of feature extraction and improve the efficiency of feature extraction.

[0092] In this implementation, the fan blade state determination model is used to normalize and perform time-frequency transformation on the plurality of first current pairs and the plurality of rotating speeds, so as to obtain current spectra and rotating speed spectra that eliminate the influence of dimensions and value ranges. The current spectra and the rotating speed spectra are used to determine the balance feature, and the extraction efficiency and accuracy of the balance feature are relatively high.

[0093] In some embodiments, the motor controller performs subsequent steps through the fan blade state determination model, normalizes the plurality of first current pairs and the plurality of rotation speeds to obtain normalized plurality of first current pairs and normalized plurality of rotation speeds. Fourier transforms the normalized plurality of first current pairs and the normalized plurality of rotation speeds to obtain the current spectrum and the rotation speed spectrum. Performs multiple convolutions and multiple full connections on the current spectrum to obtain the first spectral feature. Performs multiple convolutions and multiple full connections on the rotation speed spectrum to obtain the second spectral feature. Determines a first fusion weight corresponding to the first spectral feature and a second fusion weight corresponding to the second spectral feature, the first fusion weight being used to represent the degree of association between the first current pair and the fan blade state of the fan blade, and the second fusion weight being used to represent the degree of association between the rotation speed and the fan blade state of the fan blade. Fuses the first spectral feature and the second spectral feature using the first fusion weight and the second fusion weight to obtain the balance feature of the fan blade.

[0094] wherein the first fusion weight and the second fusion weight are set by the technician according to the actual situation, and the embodiments of the present application do not limit this. The convolution kernel used in the multiple convolutions and the full connection matrix used in the multiple full connections are both determined gradually in the process of training the fan blade state determination model.

[0095] In some embodiments, the motor controller performs subsequent steps through the fan blade state determination model, processes the plurality of first current pairs and the plurality of rotation speeds using a normalization function to obtain normalized plurality of first current pairs and normalized plurality of rotation speeds. Fast Fourier transforms the normalized plurality of first current pairs and the normalized plurality of rotation speeds to obtain the current spectrum and the rotation speed spectrum. Performs multiple convolutions and multiple full connections on the current spectrum to obtain the first spectral feature. Performs multiple convolutions and multiple full connections on the rotation speed spectrum to obtain the second spectral feature. Determines a first fusion weight and a second fusion weight. Multiplies the first fusion weight by the first spectral feature to obtain a first fusion feature. Multiplies the second fusion weight by the second spectral feature to obtain a second fusion feature. Adds the first fusion feature and the second fusion feature to obtain the balance feature of the fan blade.

[0096] The following describes another feature extraction method.

[0097] In some embodiments, the motor controller performs subsequent steps through the fan blade state determination model, normalizes the plurality of first current pairs and the plurality of rotation speeds to obtain normalized plurality of first current pairs and normalized plurality of rotation speeds. Fourier transforms the normalized plurality of first current pairs and the normalized plurality of rotation speeds to obtain the current spectrum and the rotation speed spectrum. Performs multiple convolutions and multiple full connections on the current spectrum to obtain the first spectral feature. Performs multiple convolutions and multiple full connections on the rotation speed spectrum to obtain the second spectral feature. Determines a first fusion weight corresponding to the first spectral feature and a second fusion weight corresponding to the second spectral feature, the first fusion weight being used to represent the degree of association between the first current pair and the fan blade state of the fan blade, and the second fusion weight being used to represent the degree of association between the rotation speed and the fan blade state of the fan blade. Fuses the first spectral feature and the second spectral feature using the first fusion weight and the second fusion weight to obtain the balance feature of the fan blade.

[0098] The time sequence coding refers to feature extraction by using the time sequence relationship of data. Unlike the above-mentioned feature extraction manner, this feature extraction manner is performed in the time domain.

[0099] In this embodiment, the plurality of first current pairs and the plurality of rotation speeds are coded by using the time sequence relationship of data to obtain current time sequence features and rotation speed time sequence features, so that the data change in time is fully utilized for feature extraction, and the accuracy of the final balance feature is high.

[0100] In some embodiments, the motor controller performs subsequent steps through the fan blade state determination model, sequentially encodes the plurality of first current pairs based on a gating mechanism to obtain the current time sequence features. The plurality of rotation speeds are sequentially encoded based on the gating mechanism to obtain the rotation speed time sequence features. A third fusion weight corresponding to the current time sequence features and a fourth fusion weight corresponding to the rotation speed time sequence features are determined, the third fusion weight is used to represent the degree of association between the first current pair and the fan blade state of the fan blade, and the fourth fusion weight is used to represent the degree of association between the rotation speed and the fan blade state of the fan blade. The current time sequence features and the rotation speed time sequence features are fused by using the third fusion weight and the fourth fusion weight to obtain the balance feature of the fan blade.

[0101] The gating mechanism includes a gating unit of the fan blade state determination model, and the fan blade state determination model includes a plurality of cells. One cell includes a plurality of gating units, and the plurality of gating units include an input gate, a forget gate and an output gate, which are used to control the flow of data in the cell and between cells. The parameters of each gating unit are gradually determined in the process of training the fan blade state determination model. The third fusion weight and the fourth fusion weight are set by the technician according to the actual situation, and the embodiments of the present application are not limited thereto.

[0102] In some embodiments, the motor controller determines the current time sequence feature of the plurality of first current pairs by inputting a first one of the plurality of first current pairs into a first cell of a first group of cells of the fan blade state determination model, processing the first one of the plurality of first current pairs by a plurality of gating units in the first cell, obtaining a cell state of the first cell and a hidden feature of the first one of the plurality of first current pairs, inputting the cell state of the first cell, the hidden feature of the first one of the plurality of first current pairs, and a second one of the plurality of first current pairs into a second cell of the first group of cells, processing the cell state of the first cell, the hidden feature of the first one of the plurality of first current pairs, and the second one of the plurality of first current pairs by a plurality of gating units in the second cell, obtaining a cell state of the second cell and the hidden feature of the first one of the plurality of first current pairs, and so on, until a cell state output by a last cell of the first group of cells of the fan blade state determination model is determined as the current time sequence feature of the plurality of first current pairs. Similarly, the motor controller determines the rotation speed time sequence feature of the plurality of rotation speeds by inputting a first one of the plurality of rotation speeds into a first cell of a second group of cells of the fan blade state determination model, processing the first one of the plurality of rotation speeds by a plurality of gating units in the first cell, obtaining a cell state of the first cell and a hidden feature of the first one of the plurality of rotation speeds, inputting the cell state of the first cell, the hidden feature of the first one of the plurality of rotation speeds, and a second one of the plurality of rotation speeds into a second cell of the second group of cells, processing the cell state of the first cell, the hidden feature of the first one of the plurality of rotation speeds, and the second one of the plurality of rotation speeds by a plurality of gating units in the second cell, obtaining a cell state of the second cell and the hidden feature of the first one of the plurality of rotation speeds, and so on, until a cell state output by a last cell of the second group of cells of the fan blade state determination model is determined as the rotation speed time sequence feature of the plurality of rotation speeds. The motor controller determines a third fusion weight and a fourth fusion weight, multiplies the third fusion weight with the current time sequence feature to obtain a third fusion feature, multiplies the fourth fusion weight with the rotation speed time sequence feature to obtain a fourth fusion feature, and adds the third fusion feature and the fourth fusion feature to obtain the balance feature of the fan blade.

[0103] 303、In a case where the fan blade state of the fan blade is balanced, the motor controller controls the motor to continue rotating according to the current control parameter.

[0104] The current control parameter refers to a parameter determined by a conventional control logic for controlling the motor, and the control manner provided in step 303 means that the rotation of the motor is not interfered.

[0105] 304、In the case that the state of the fan blade is unbalanced, the motor controller controls the motor to reduce the rotating speed or stop rotating.

[0106] In some embodiments, in the case that the state of the fan blade is unbalanced, the motor controller determines the number of times that the state of the fan blade is continuously determined to be unbalanced. In the case that the number of times is greater than or equal to a first preset number of times, the motor controller controls the motor to reduce the rotating speed. In the case that the number of times is greater than or equal to a second preset number of times, the motor controller controls the motor to stop rotating, the second preset number of times being greater than the first preset number of times.

[0107] The first preset number of times and the second preset number of times are set by technicians according to actual conditions, and embodiments of the present application do not limit this. The first preset number of times and the second preset number of times are configured to avoid the situation that the outdoor unit is incorrectly stopped due to an error in single fan blade state identification.

[0108] In this implementation, in the case that the state of the fan blade is unbalanced, the number of times that the state of the fan blade is continuously determined to be unbalanced is used to control the motor to reduce the rotating speed or stop rotating, thereby avoiding the situation that the outdoor unit is incorrectly stopped due to an error in single fan blade state identification, and affecting the refrigeration or heating effect.

[0109] In some embodiments, in the case that the state of the fan blade is unbalanced, the motor controller sends first alarm information to an associated device of the outdoor unit, the first alarm information being used to indicate that the fan blade of the outdoor unit may be in an unbalanced state. In addition, in the case that the number of times is greater than or equal to the first preset number of times, the motor controller sends second alarm information to the associated device of the outdoor unit, the second alarm information being used to indicate that the fan blade of the outdoor unit is in an unbalanced state and the motor has been reduced in speed. Alternatively, in the case that the number of times is greater than or equal to the second preset number of times, the motor controller sends third alarm information to the associated device of the outdoor unit, the third alarm information being used to indicate that the fan blade of the outdoor unit is in an unbalanced state and the motor has been stopped.

[0110] The associated device is a device used by a maintenance personnel of the outdoor unit, and the associated device has a binding relationship with the outdoor unit.

[0111] By sending the first alarm information, the second alarm information, and the third alarm information, the user can be reminded to timely process the fan blade to eliminate the unbalanced state.

[0112] For the technical solutions provided in the embodiments of the present application, the above steps 301-304 will be described below in conjunction with FIG. 5. Referring to FIG. 5, the ejector air chamber outdoor unit is started, and a set of input currents and a set of rotation data are obtained. The set of input currents and the set of rotation data are input into a fan blade state determination model. The fan blade state determination model is used to convert the set of input currents based on a plurality of rotation positions in the set of rotation data, to obtain a plurality of first current pairs. The plurality of first current pairs and a plurality of rotation speeds in the set of rotation data are normalized and time-frequency transformed to obtain a current spectrum and a rotation speed spectrum. The current spectrum and the rotation speed spectrum are input into the fan blade state determination model, and the fan blade state determination model is used to determine the fan blade state of the fan blade based on the current spectrum and the rotation speed spectrum. In the case where the fan blade state of the fan blade is balanced, the motor continues to rotate. In the case where the fan blade state of the fan blade is unbalanced, the motor reduces the rotation speed or stops rotating.

[0113] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described one by one here.

[0114] Through the technical solutions provided in the embodiments of the present application, a plurality of input current data of a motor driving a fan blade of an ejector air chamber outdoor unit and a plurality of rotation data are obtained. The plurality of input current data and the plurality of rotation data are processed by a fan blade state determination model to determine the fan blade state of the fan blade, so as to realize the identification of the fan blade state of the fan blade. The motor of the ejector air chamber outdoor unit is controlled based on the fan blade state, so as to reduce the probability of damage of the fan blade and the motor due to unbalanced fan blade.

[0115] The training method of the fan blade state determination model provided in the embodiments of the present application will be described below. Taking an electronic device as an example, the method includes the following steps.

[0116] 601. The electronic device obtains a plurality of sample data sets of the ejector air chamber outdoor unit and a labeled fan blade state corresponding to each sample data set. The sample data set includes a sample input current data set and a sample rotation data set.

[0117] The sample input current data set includes multiple sample input current data, and the sample input current data is historical input current data collected by the top-out air conditioner during operation, which is the actual input current data of the motor of the top-out air conditioner. Correspondingly, the sample rotation data set includes multiple sample rotation data, and the sample rotation data is historical rotation data collected by the top-out air conditioner during operation, which is the actual rotation data of the motor of the top-out air conditioner. In addition, for a sample data set, the sample input current data and the sample rotation data in the sample data set exist in pairs, that is, each sample input current data has corresponding sample rotation data, and the collection time is the same or similar.

[0118] In some embodiments, the electronic device obtains multiple first sample data sets of the top-out air conditioner, the multiple first sample data sets are collected when the fan blade state of the fan blade is balanced, and different first sample data sets are collected when the top-out air conditioner is in different back air blocking ratios. The electronic device obtains multiple second sample data sets of the top-out air conditioner, the multiple second sample data sets are sample data sets collected when the fan blade state of the fan blade is unbalanced, and different second sample data sets are collected when the top-out air conditioner is in different unbalanced loads.

[0119] In some embodiments, the back air blocking ratio includes 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90%, etc., and the present embodiment is not limited thereto. The unbalanced load is realized by placing a weight on the fan blade of the top-out air conditioner by the technician, which is to simulate the working condition of the top-out air conditioner under different fan blade states. In the present embodiment, the unbalanced load is represented by the mass of the weight, and in some embodiments, the unbalanced load includes 5g, 10g, and 20g, etc., and the present embodiment is not limited thereto. The multiple first sample data sets can be regarded as positive samples for training the fan blade state determination model, and the multiple second sample data sets can be regarded as negative samples for training the fan blade state determination model. By training the fan blade state determination model with positive samples and negative samples, the fan blade state determination model can learn the knowledge related to the balanced and unbalanced fan blade states, thereby realizing the binary classification of balanced and unbalanced. Collecting the first sample data set under different back air blocking ratios and collecting the second sample data set under different balanced loads is to cover as many working scenarios of the top-out air conditioner as possible to improve the generalization ability of the fan blade state determination model trained.

[0120] 602、The electronic device inputs the plurality of sample data sets into the fan blade state determination model, processes each sample data set through the fan blade state determination model, and obtains a predicted fan blade state corresponding to each sample data set.

[0121] The step 602 belongs to the same inventive concept as the step 302 described above. For details, refer to the description of the step 302 above.

[0122] 603、The electronic device trains the fan blade state determination model based on difference information between the predicted fan blade state corresponding to each sample data set and the labeled fan blade state.

[0123] In some embodiments, in any round of training, the electronic device trains the fan blade state determination model based on first difference information between the predicted fan blade state corresponding to the first sample data set used in the round and the labeled fan blade state, and second difference information between the predicted fan blade state corresponding to the second sample data set used in the round and the labeled fan blade state.

[0124] In some embodiments, in any round of training, the electronic device substitutes the first difference information and the second difference information into a loss function of the fan blade state determination model to determine a loss value of the round. The electronic device uses the loss value to perform back propagation in the fan blade state determination model using the gradient descent method to adjust model parameters of the fan blade state determination model, thereby realizing one round of training of the fan blade state determination model.

[0125] In some embodiments, the loss function is a cross-entropy loss function. In any round of training, the electronic device substitutes the first difference information and the second difference information into the cross-entropy loss function of the fan blade state determination model to obtain the loss value of the round. The electronic device uses an Adam optimizer to perform back propagation in the fan blade state determination model based on the loss value to adjust the model parameters of the fan blade state determination model, thereby realizing one round of training of the fan blade state determination model.

[0126] The Adam (Adaptive Moment Estimation) optimizer is an adaptive learning rate optimization algorithm. It combines the ideas of momentum and adaptive learning rate, and adjusts the learning rate by exponentially weighted moving average of the first moment estimate and the second moment estimate of the gradient.

[0127] To more clearly illustrate the above model training process, the principle of the fan blade state determination model is described below, taking the case that the fan blade state determination model is trained based on the principle of a support vector machine.

[0128] Referring to FIG. 7, the current spectrum (processed data A) and the rotating speed spectrum (processed data B) are regarded as a data point, and the corresponding data point is white when the fan blade state of the fan blade is balanced and the corresponding data point is black when the fan blade state of the fan blade is unbalanced, wherein the processing includes normalization and time-frequency transformation. Assuming that the two groups of data points (one group of white data points and one group of black data points) are linearly separable, there must be a hyperplane y = w * x + b = 0 that can completely separate the two groups of data points, and in order to make the hyperplane more robust, the best hyperplane, that is, the hyperplane that can separate the two groups of data points with the largest interval, is found, and in some embodiments, the best hyperplane is also referred to as the maximum interval hyperplane, and the goal of training the fan blade state determination model is to find this hyperplane, thereby realizing the binary classification of the data points. In addition, some data points in the sample that are closest to the hyperplane are referred to as support vectors.

[0129] wherein the maximum interval hyperplane should satisfy the following two conditions: ① the two groups of data points are respectively distributed on the two sides of the hyperplane; and ② the distance of the data points closest to the hyperplane on the two sides to the hyperplane should be the largest.

[0130] Suppose that there is a hyperplane y = w * x + b = 0, and for any data point in space, the distance of the data point to the hyperplane is:

[0131] |w*x+b| / ||w||

[0132] It is specified that the distance of the support vector to the hyperplane is d, and thus the following relationship is obtained:

[0133] (w*x+b) / ||w||≥d,y=1

[0134] (w*x+b) / ||w||≤-d,y=-1

[0135] The two equations are combined and can be simply written as y * (w * x + b) ≥ 1

[0136] Thus, the condition is a constraint condition for satisfying linear separability, and at the same time, in the case of guaranteeing the constraint condition, the distance d should be as large as possible to make the interval between the two data points larger, that is, the following optimization problem is satisfied:

[0137] max (d) s.t. yi * (w * xi + b) ≥ 1, d = |w * x + b| / ||w||

[0138] If the data satisfies linear separability, there must be a maximum interval hyperplane that can completely separate the data points, and after the above formula is derived and simplified, the following optimization problem is obtained:

[0139] min (1 / 2 * ||w||^2) s.t. yi * (w * xi + b) ≥ 1

[0140] For the convex quadratic programming problem of 1 / 2*||w||^2 s.t. yi*(w*xi+b)≥1, the optimal problem of dual variables is obtained by Lagrange duality transformation, and the optimal solution of the original problem is obtained by the equivalence of the two problems. The Lagrange duality transformation is to add the Lagrange operator α to the constraint condition, and define the Lagrange function as follows:

[0141] L(w,b,α)=1 / 2*||w||^2-∑αi*(yi*(w*xi+b)-1)

[0142] Let θ(w)=max(αi>0)(L(w,b,α)), then the minimization of 1 / 2*||w||^2 in the original problem is equivalent to the direct minimization of θ(w), and the objective function is transformed into:

[0143] min(w,b)(θ(w))=min(w,b)max(αi>0)(L(w,b,α))

[0144] The maximum minimum problem can be interchanged, and the problem is finally changed into:

[0145] max(αi>0)min(w,b)(L(w,b,α))

[0146] Solving the problem can obtain the maximum interval hyperplane.

[0147] However, in reality, most classification problems are not linearly separable, and for non-linearly separable problems, there is no such hyperplane. For this case, as shown in FIG. 8, the solution is to map the low-dimensional linearly inseparable samples to a high-dimensional space, so that the sample points are linearly separable in the high-dimensional space. For samples that are linearly inseparable in a finite-dimensional vector space, they are mapped to a higher-dimensional vector space, and then learned by the maximum interval method to obtain a support vector machine, which is a nonlinear SVM. The process of mapping to a higher-dimensional vector space, that is, the feature extraction process in the embodiment of the application. Let φ(x) represent the mapping of x to the new feature space to the new vector, and the hyperplane is y=w*φ(x)+b=0.

[0148] After mapping the low-dimensional space to the high-dimensional space, the dimension will be very large, and if all the sample points are multiplied and calculated, the calculation amount will be too large, so a kernel function k(xi,xj)=(φ(xi),φ(xj)) is used, which is equal to the inner product of xi and xj in the feature space, and the result calculated by the function k(xi,xj) in the original sample space. Then the inner product of the high-dimensional or infinite-dimensional space is no longer needed.

[0149] Therefore, as shown in FIG. 9, a Gaussian kernel function is selected as a mapping form of the data, and an expression of the Gaussian kernel function is as follows:

[0150] k(xi,xj)=exp(-(||xi-xj|| / 2δ^2))

[0151] Based on the kernel function, a support vector machine-based fan blade state determination model is trained by continuously optimizing model parameters through the processed sample data set.

[0152] FIG. 10 is a structural schematic diagram of a control device of a motor according to an embodiment of the present application. As shown in FIG. 10, the device includes a data acquisition module 1001, a fan blade state determination module 1002, and a control module 1003.

[0153] The data acquisition module 1001 is configured to acquire an input current data set and a rotation data set of a motor of an ejection air chamber outdoor unit. The motor is configured to drive a fan blade of the ejection air chamber outdoor unit to rotate. The input current data set includes a plurality of input current data, and the rotation data set includes a plurality of rotation data.

[0154] The fan blade state determination module 1002 is configured to input the input current data set and the rotation data set into a fan blade state determination model, process the input current data set and the rotation data set through the fan blade state determination model, and obtain a fan blade state of the fan blade. The fan blade state includes balance and imbalance.

[0155] The control module 1003 is configured to control the motor based on the fan blade state of the fan blade.

[0156] It should be noted that the control device of the motor provided in the above embodiments is only used as an example to divide the above functional modules when controlling the motor. In actual applications, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the ejection air chamber outdoor unit is divided into different functional modules to complete all or part of the above described functions. In addition, the control device of the motor and the control method of the motor provided in the above embodiments belong to the same concept, and the specific implementation process is described in detail in the method embodiments, which will not be described here.

[0157] According to the technical solutions provided in the embodiments of the present application, a plurality of input current data and a plurality of rotation data of a motor driving a fan blade of an ejection air chamber outdoor unit are acquired. The plurality of input current data and the plurality of rotation data are processed by using a fan blade state determination model to determine a fan blade state of the fan blade, so as to realize identification of the fan blade state of the fan blade. The motor of the ejection air chamber outdoor unit is controlled based on the fan blade state, so as to reduce the probability of damage of the fan blade and the motor due to imbalance of the fan blade.

[0158] FIG. 11 is a structural schematic diagram of a top-out air chamber outdoor unit according to an embodiment of the present application. The top-out air chamber outdoor unit 1100 includes one or more processors 1101 and one or more memories 1102.

[0159] The processor 1101 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 1101 can be implemented in at least one of a hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), a PLA (Programmable Logic Array). The processor 1101 can also include a main processor and a coprocessor. The main processor is a processor configured to process data in a wake-up state, also referred to as a CPU (Central Processing Unit). The coprocessor is a low-power processor configured to process data in a standby state. In some embodiments, the processor 1101 can be integrated with a GPU (Graphics Processing Unit) configured to be responsible for rendering and drawing of content required to be displayed by a display screen. In some embodiments, the processor 1101 can further include an AI (Artificial Intelligence) processor configured to process computing operations related to machine learning.

[0160] The memory 1102 can include one or more computer-readable storage media, which can be non-transitory. The memory 1102 can also include a high-speed random access memory, and a nonvolatile memory, in some embodiments, one or more disk storage devices, flash storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 1102 is configured to store at least one computer program for being executed by the processor 1101 to implement the control method of the motor provided by the method embodiment of the present application.

[0161] Those skilled in the art can understand that the structure shown in FIG. 11 does not constitute a limitation on the top-out air chamber outdoor unit 1100, and can include more or fewer components than those shown, or combine certain components, or adopt different component arrangements.

[0162] In the example embodiment, a computer readable storage medium, for example, a memory including a computer program executable by a processor to implement the control method of the motor in the above embodiment is also provided. For example, the computer readable storage medium can be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0163] In the example embodiment, a computer program product or computer program including program code stored in a computer readable storage medium is also provided, and the processor of the computer device reads the program code from the computer readable storage medium, and the processor executes the program code to make the computer device implement the control method of the motor.

[0164] In some embodiments, the computer program related to the embodiments of the present application can be deployed to execute on one computer device, or on multiple computer devices located in one place, or on multiple computer devices distributed in multiple places and interconnected through a communication network, which can constitute a blockchain system.

[0165] Those of ordinary skill in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by a program instructing relevant hardware, and the program can be stored in a computer readable storage medium, and the storage medium mentioned above can be a Read-Only Memory, a magnetic disk or an optical disk, etc.

[0166] The above is only an optional embodiment of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A control method of an electric machine, the method comprising: obtaining a set of input current data and a set of rotation data of an electric machine of an ejecting air handling unit, the electric machine being configured to drive an impeller of the ejecting air handling unit to rotate, the set of input current data comprising a plurality of input current data, the set of rotation data comprising a plurality of rotation data; inputting the set of input current data and the set of rotation data into an impeller state determination model, processing the set of input current data and the set of rotation data by the impeller state determination model to obtain an impeller state of the impeller, the impeller state comprising balanced and unbalanced; and controlling the electric machine based on the impeller state of the impeller. the rotation data comprising rotation speeds and rotation positions of a rotor of the electric machine, the processing the set of input current data and the set of rotation data by the impeller state determination model to obtain the impeller state of the impeller comprising:

2. The method of claim 1, wherein, transforming, by the impeller state determination model, corresponding input current data in the set of input current data based on a plurality of rotation positions in the set of rotation data to obtain a plurality of first current pairs of the electric machine, one of the first current pairs comprising a set of d-axis current and q-axis current; and determining, by the impeller state determination model, the impeller state of the impeller based on the plurality of first current pairs of the electric machine and a plurality of rotation speeds in the set of rotation data. the transforming, by the impeller state determination model, corresponding input current data in the set of input current data based on a plurality of rotation positions in the set of rotation data to obtain a plurality of first current pairs of the electric machine comprising:

3. The method of claim 2, wherein, performing, by the impeller state determination model, Clarke transformation on each of the input current data in the set of input current data to obtain a plurality of second current pairs, one of the second current pairs comprising a set of a-axis current and b-axis current in a two-phase static coordinate system; and performing, by the impeller state determination model, Park transformation on corresponding second current pairs in the plurality of second current pairs based on a plurality of rotation positions in the set of rotation data to obtain the plurality of first current pairs of the electric machine. the determining, by the impeller state determination model, the impeller state of the impeller based on the plurality of first current pairs of the electric machine and a plurality of rotation speeds in the set of rotation data comprising:

4. The method of claim 2, wherein, performing, by the impeller state determination model, feature extraction on the plurality of first current pairs and the plurality of rotation speeds in the set of rotation data to obtain a balanced feature of the impeller; performing, by the impeller state determination model, full connection and normalization on the balanced feature to obtain an impeller state prediction score of the impeller; in a case where the impeller state prediction score is greater than or equal to a state score threshold, determining the impeller state of the impeller as balanced; and in a case where the impeller state prediction score is less than the state score threshold, determining the impeller state of the impeller as unbalanced. ​ 5. The method of claim 4, wherein, The balance feature of the fan blade is obtained by performing feature extraction on the multiple first current pairs and the multiple rotating speeds in the rotating data set through the fan blade state determination model, including: The current spectrum of the multiple first current pairs and the rotating speed spectrum of the multiple rotating speeds are obtained by performing normalization and time-frequency transformation on the multiple first current pairs and the multiple rotating speeds in the rotating data set through the fan blade state determination model; The first spectrum feature of the rotating speed spectrum and the second spectrum feature of the rotating speed spectrum are obtained by performing feature extraction on the current spectrum and the rotating speed spectrum through the fan blade state determination model; and The balance feature of the fan blade is obtained by fusing the first spectrum feature and the second spectrum feature through the fan blade state determination model.

6. The method of claim 5, wherein, The balance feature of the fan blade is obtained by fusing the first spectrum feature and the second spectrum feature through the fan blade state determination model, including: The first fusion weight corresponding to the first spectrum feature and the second fusion weight corresponding to the second spectrum feature are determined through the fan blade state determination model, the first fusion weight is configured to represent the degree of association between the first current pair and the fan blade state of the fan blade, and the second fusion weight is configured to represent the degree of association between the rotating speed and the fan blade state of the fan blade; and The first spectrum feature and the second spectrum feature are fused by using the first fusion weight and the second fusion weight to obtain the balance feature of the fan blade.

7. The method of claim 4, wherein, The balance feature of the fan blade is obtained by performing feature extraction on the multiple first current pairs and the multiple rotating speeds in the rotating data set through the fan blade state determination model, including: The current time sequence feature of the multiple first currents and the rotating speed time sequence feature of the multiple rotating speeds are obtained by performing time sequence coding on the multiple first current pairs and the multiple rotating speeds through the fan blade state determination model; and The balance feature of the fan blade is obtained by fusing the current time sequence feature and the rotating speed time sequence feature through the fan blade state determination model.

8. The method of any one of claims 1-7, wherein, The motor is controlled based on the fan blade state of the fan blade, including: In the case that the fan blade state of the fan blade is balanced, the motor is controlled to continue rotating according to the current control parameter; and In the case that the fan blade state of the fan blade is unbalanced, the motor is controlled to reduce the rotating speed or stop rotating.

9. The method of claim 8, wherein, In the case that the fan blade state of the fan blade is unbalanced, the motor is controlled to reduce the rotating speed or stop rotating, including: In the case that the fan blade state of the fan blade is unbalanced, the number of times that the fan blade state is continuously determined to be unbalanced is determined; In the case that the number of times is greater than or equal to a first preset number of times, the motor is controlled to reduce the rotating speed; and In the case that the number of times is greater than or equal to a second preset number of times, the motor is controlled to stop rotating, the second preset number of times is greater than the first preset number of times.

10. The method of any one of claims 1-9, wherein, The training method of the fan blade state determination model includes: obtain a plurality of sample data sets of the ejecting air outdoor unit and a labeled blade state corresponding to each of the sample data sets, the sample data set including a sample input current data set and a sample rotation data set; input the plurality of sample data sets into the blade state determination model, process each of the sample data sets through the blade state determination model, and obtain a predicted blade state corresponding to each of the sample data sets; and train the blade state determination model based on difference information between the predicted blade state and the labeled blade state corresponding to each of the sample data sets.

11. The method of claim 10, wherein, The obtaining the plurality of sample data sets of the ejecting air outdoor unit and the labeled blade state corresponding to each of the sample data sets includes: obtaining a plurality of first sample data sets of the ejecting air outdoor unit, the plurality of first sample data sets being collected when the blade state of the blade is balanced, and different first sample data sets being collected when the ejecting air outdoor unit is in different back air blocking ratios; and obtaining a plurality of second sample data sets of the ejecting air outdoor unit, the plurality of second sample data sets being collected when the blade state of the blade is unbalanced, and different second sample data sets being collected when the ejecting air outdoor unit is in different unbalanced loads.

12. A control device of an electric machine, the device comprising: a data obtaining module configured to obtain an input current data set and a rotation data set of an electric machine of an ejecting air outdoor unit, the electric machine being configured to drive a blade of the ejecting air outdoor unit to rotate, the input current data set including a plurality of input current data, and the rotation data set including a plurality of rotation data; a blade state determination module configured to input the input current data set and the rotation data set into a blade state determination model, process the input current data set and the rotation data set through the blade state determination model, and obtain a blade state of the blade, the blade state including balance and imbalance; and a control module configured to control the electric machine based on the blade state of the blade.

13. An ejecting air outdoor unit, the ejecting air outdoor unit comprising one or more processors and one or more memories, the one or more memories having stored therein at least one computer program, the computer program being loaded and executed by the one or more processors to implement the control method of the electric machine according to any one of claims 1 to 11.

14. A computer-readable storage medium having stored therein at least one computer program, the computer program being loaded and executed by a processor to implement the control method of the electric machine according to any one of claims 1 to 11. ​

Images