Calculation system, magnetic bearing system, and refrigeration device

Abstract

A calculation system 100 comprises: a signal generating unit 30 that causes an excitation signal for exciting a magnetic bearing 20 to be input from a magnetic bearing control device 10 to the magnetic bearing 20; and a calculating unit 512 that calculates a frequency response of part or all of the magnetic bearing control device 10 and the magnetic bearing 20 when the excitation signal is input to the magnetic bearing 20. The excitation signal is a split multi-sine signal having, on different time axes, a first waveform obtained by superimposing a plurality of sine waves having different frequencies, and a second waveform obtained by superimposing a plurality of sine waves having different frequencies and having a waveform different from the first waveform.

Description

Calculation system, magnetic bearing system, and refrigeration device

[0001] This disclosure relates to a calculation system, a magnetic bearing system, and a refrigeration system.

[0002] When calculating the frequency response (transfer function) of a control device or controlled object, the frequency of the excitation signal input to the control device and controlled object is typically changed sequentially, and the frequency responses at multiple frequencies are calculated. In this case, since each frequency is changed in series on the time axis, the time required for calculation becomes long.

[0003] Therefore, Patent Document 1 discloses a servo analyzer that applies a broadband signal from the servo analyzer's signal source to the system under test, determines the spectrum of the measurement frequency band based on the applied broadband signal and the output of the system under test, and measures the transfer function. The broadband signal includes a multisine signal.

[0004] Furthermore, Patent Document 2 also discloses a frequency response device that generates a multisine signal consisting of sine waves of multiple frequencies and inputs it to the controlled object (object under measurement), and calculates the frequency characteristics of the controlled object from the sampling data of the input signal and the sampling data of the output signal. In this way, conventionally, a multisine signal with multiple frequencies superimposed is output all at once, and the input signal and output signal corresponding to the multisine signal are analyzed in a computing device to shorten the time required for calculation.

[0005] Japanese Unexamined Patent Publication No. 8-94690 Patent No. 7007318

[0006] By the way, when a multisine signal is used as the signal applied to the controlled object, if the frequency bandwidth to be measured is wide, or if the number of frequencies is large, it is necessary to use a large computer or similar as the processing unit, which can perform a large number of calculations per unit time or has a large memory capacity.

[0007] On the other hand, depending on the controlled system, there is a need to measure the frequency response at each usage site. Bringing large computing devices to each of multiple usage sites to measure the frequency response would require time and cost for arranging, transporting, and installing these devices, which would deviate from the objective of measuring the frequency response in a short time.

[0008] This disclosure provides a technology that enables obtaining good frequency characteristics of a controlled object even at the site where the controlled object is used.

[0009] According to a first aspect of the present disclosure, the device includes a signal generation unit that causes a control device to input an excitation signal to a controlled object for causing the controlled object to vibrate, and a calculation unit that calculates the frequency response of part or all of the control device and the controlled object when the excitation signal is input to the controlled object, wherein the excitation signal is a divided multisine signal having a first waveform obtained by superimposing a plurality of sine waves of different frequencies, and a second waveform obtained by superimposing a plurality of sine waves of different frequencies, and having a waveform different from the first waveform, on different time axes.

[0010] As described above, the calculation system, when calculating the frequency response, uses a divided multisine signal as the excitation signal, allowing it to input multiple waveforms with an appropriate number of sine waves to the controlled object, shifted on the time axis according to the performance of the computing device. This makes it possible to efficiently obtain the frequency response of the control device and the controlled object according to the performance of the computing device. For example, the calculation system can obtain good frequency characteristics of the controlled object even in usage sites where it is difficult to apply a high-performance computing device.

[0011] Furthermore, the signal generation unit switches all of the multiple sine waves forming the first waveform and all of the multiple sine waves forming the second waveform at the same timing.

[0012] In this way, by switching all of the multiple sine waves of the first waveform and all of the multiple sine waves of the second waveform at the same timing, the calculation system can easily manage the multiple sine waves used to calculate the frequency response.

[0013] Furthermore, the signal generation unit switches from the first waveform to the second waveform by switching some of the multiple sine waves forming the first waveform to sine waves of different frequencies.

[0014] In this way, by switching some of the sine waves among the multiple sine waves forming the first waveform to sine waves of different frequencies, the calculation system can more efficiently calculate the frequency response of all the sine waves.

[0015] Furthermore, the number of sine waves forming the first waveform is the same as the number of sine waves forming the second waveform.

[0016] Thus, by having the same number of sine waves forming the first waveform as the same number of sine waves forming the second waveform, the calculation system can calculate the frequency response of all sine waves more efficiently.

[0017] Furthermore, the signal generation unit sets the amplitude of the multiple sine waves in the divided multisine signal to be larger for sine waves with higher frequencies.

[0018] This allows the calculation system to accurately calculate the frequency response even for high-frequency sine waves.

[0019] Furthermore, the frequency response includes the characteristics of the amplitude ratio and phase difference between the input signal and the output signal, and the calculation unit calculates the amplitude ratio or phase difference for each of the multiple sine waves by performing a discrete Fourier transform on the input signal and the output signal acquired from the control device or the controlled object.

[0020] In this way, by performing a discrete Fourier transform, the calculation system can easily calculate the frequency response for each sine wave at multiple frequencies.

[0021] Furthermore, the calculation unit calculates the amplitude ratio or phase difference for each of the multiple sine waves forming the first waveform by adding up the values ​​calculated for each arbitrary time point of the first waveform, and calculates the amplitude ratio or phase difference for each of the multiple sine waves forming the second waveform by adding up the values ​​calculated for each arbitrary time point of the second waveform.

[0022] Thus, in the discrete Fourier transform, the amplitude ratio or phase difference of the frequency response can be calculated independently of each other by summing the values ​​calculated at any given time in the first and second waveforms. As a result, the data used can be immediately erased, and the increase in the memory capacity of the arithmetic unit can be suppressed.

[0023] Furthermore, the signal generation unit is provided inside the control device.

[0024] By integrating the signal generation unit into the control device in this way, the need to prepare a separate signal generation unit is eliminated, making it easier to obtain the frequency response.

[0025] Furthermore, the signal generation unit is located outside the control device.

[0026] By placing the signal generation unit outside the control device in this way, it is possible to suppress the increase in size of the control device.

[0027] Furthermore, the system has a user setting unit that allows the user to set the frequency band or the number of frequencies of the excitation signal input to the controlled object, and the signal generation unit automatically sets the number of sine waves included in the first and second waveforms of the divided multisine signal based on the frequency band or the number of frequencies set by the user setting unit.

[0028] This allows the calculation system to suppress the user-defined load when acquiring the frequency response, while also enabling the appropriate generation of segmented multisine signals.

[0029] Furthermore, a second aspect of the present disclosure is a magnetic bearing system comprising the above-described calculation system, a magnetic bearing that non-contactually supports a rotating shaft, and a magnetic bearing control device for controlling the magnetic bearing, wherein the controlled object is the magnetic bearing and the control device is a magnetic bearing control device.

[0030] This allows the calculation system to reliably measure the frequency response of the magnetic bearing system.

[0031] Further, a third aspect of the present disclosure is a refrigeration device including a compressor having a magnetic bearing as a control target, a control device that controls the operation of the magnetic bearing, and a refrigerant circuit provided with the compressor and circulating refrigerant based on the operation of the compressor. The control device inputs a vibration signal for vibrating the magnetic bearing to the magnetic bearing, and causes a calculation unit to calculate the frequency response of part or all of the magnetic bearing and the control device when the vibration signal is input to the magnetic bearing. The vibration signal is a divided multi-sine signal having, on different time axes, a first waveform obtained by superimposing a plurality of sine waves having different frequencies, and a second waveform obtained by superimposing a plurality of sine waves having different frequencies and having a waveform different from the first waveform.

[0032] Even in this case, the refrigeration device can obtain good frequency characteristics of the magnetic bearing even at the place where the compressor is used.

[0033] It is a diagram schematically showing the overall configuration of an air conditioner which is a refrigeration device according to an embodiment. It is a diagram showing the overall configuration of a calculation system according to an embodiment. It is a diagram showing an example of generation of a multi-sine signal. It is an explanatory diagram showing a divided multi-sine signal according to a first example. It is an explanatory diagram showing a divided multi-sine signal according to a second example. It is an explanatory diagram showing a divided multi-sine signal according to a third example. It is a flowchart showing a method for calculating a frequency response. It is an explanatory diagram showing a state in which a frequency response is calculated by discrete Fourier transform of an arithmetic unit. FIG. 9(A) is an explanatory diagram showing an example of performing discrete Fourier transform on a time series waveform. FIG. 9(B) is an explanatory diagram showing the relationship between the calculation formula of discrete Fourier transform and the frequency response of each frequency. It is a diagram showing the overall configuration of a calculation system according to a modified example.

[0034] Hereinafter, embodiments for implementing the present disclosure will be described with reference to the drawings. In each drawing, the same reference numerals are given to the same components, and duplicate explanations may be omitted. Further, each drawing may exaggerate or simplify dimensions, ratios or numbers as necessary for easy understanding of the invention.

[0035] <Configuration of Refrigeration Device> First, an air conditioner 1, which is a refrigeration device to which the calculation system 100 according to the present disclosure is applied, will be described with reference to FIG. 1. The air conditioner 1 is a device that performs air conditioning of a living space by a vapor compression refrigeration cycle. The air conditioner 1 according to the embodiment includes a compressor 2, a heat source side heat exchanger 3, an expansion mechanism 4, and a utilization side heat exchanger 5, and executes a cooling operation.

[0036] For example, the compressor 2, the heat source side heat exchanger 3, and the expansion mechanism 4 are provided inside the heat source side unit of the air conditioner 1. The heat source side unit is an outdoor unit disposed outside the building in which the air conditioner 1 is installed. On the other hand, the utilization side heat exchanger 5 is provided inside the utilization side unit of the air conditioner 1. The utilization side unit is an indoor unit installed in the living space of the building in which the air conditioner 1 is installed. Note that the expansion mechanism 4 may be provided not only in the heat source side unit but also in the utilization side unit.

[0037] The heat source side unit and the utilization side unit are connected to each other via a first pipe 6 and a second pipe 7. The compressor 2, the heat source side heat exchanger 3, the expansion mechanism 4, the utilization side heat exchanger 5, the first pipe 6, and the second pipe 7 are sealed inside, forming a refrigerant circuit 8 through which the refrigerant can circulate.

[0038] A single-stage centrifugal compressor can be applied to the compressor 2. The compressor 2 sucks in a low-pressure refrigerant flowing through the second pipe 7, compresses this refrigerant into a high-pressure refrigerant, and discharges it to the heat source side heat exchanger 3. For example, the compressor 2 has a casing (not shown) and includes a magnetic bearing 20, a rotating shaft 25, a compression part 26, a motor 27, etc. inside this casing.

[0039] The magnetic bearing 20 levitates the rotating shaft 25 and axially supports the rotating shaft 25 in a non-contact and rotatable manner. In other words, the magnetic bearing 20 supports the load in the radial direction of the rotating shaft 25. Note that in addition to this magnetic bearing 20, the compressor 2 may include a magnetic bearing that supports the load in the axial direction (thrust direction) of the rotating shaft 25.

[0040] Specifically, the magnetic bearing 20 has a plurality of electromagnets arranged at intervals in the circumferential direction. The magnetic bearing 20 generates a magnetic field between each electromagnet and the rotating shaft 25 by supplying a current controlled by an inverter to the coil of each electromagnet, thereby magnetically levitating and supporting the rotating shaft 25 without contact. The inverter of the magnetic bearing 20 is controlled by a magnetic bearing control device 10 connected to the magnetic bearing 20.

[0041] The magnetic bearing control device 10 is a computer or microcontroller that has a control circuit board including a processor, memory, input / output interface, and communication interface (not shown), and controls only the magnetic bearing 20. The magnetic bearing control device 10 may also be installed in conjunction with the control unit that controls the entire heat source side unit of the air conditioning system 1.

[0042] Furthermore, the magnetic bearing 20 is equipped with a gap sensor 28 that detects the radial gap between the rotating shaft 25 and each electromagnet. The type of gap sensor 28 is not particularly limited, but for example, a displacement sensor capable of detecting the radial position of the rotating shaft 25 can be used. The magnetic bearing control device 10 controls the radial position of the rotating shaft 25 by controlling the power supplied to the coils of each electromagnet based on the information of the actual position (radial gap) of the rotating shaft 25 detected by the gap sensor 28.

[0043] The compression section 26 of the compressor 2 is the part that compresses the refrigerant drawn in from the second pipe 7, and is located in the compression space of the casing. The compression section 26 is mainly composed of an impeller provided at one axial end of the rotating shaft 25. The impeller rotates based on the rotational drive of the motor 27, drawing in and compressing the refrigerant from the second pipe 7 to a high pressure, and then discharges the refrigerant toward the heat exchanger 5 on the utilization side.

[0044] Furthermore, the motor 27 of the compressor 2 is located at the other axial end of the rotating shaft 25, opposite to the impeller. The motor 27 has a rotor and stator (not shown) that rotate the rotating shaft 25, and its rotation is controlled by the control unit (not shown) of the air conditioning system 1. The mechanisms and arrangement of the magnetic bearing 20, thrust bearing, motor 27, etc. of the compressor 2 are not limited to the above and can take various configurations. For example, the compressor 2 may have the magnetic bearing of the thrust shaft located at the other axial end opposite to the impeller, while the motor 27 is located on the impeller side.

[0045] The compressor 2 supports the rotating shaft 25 non-contact with the magnetic bearing 20 as described above, allowing the impeller to rotate smoothly. As a result, the compressor 2 can stably circulate the refrigerant to the refrigerant circuit 8 while compressing the refrigerant.

[0046] On the other hand, the heat source side heat exchanger 3 of the heat source side unit receives compressed refrigerant, performs heat exchange between the refrigerant and air or water, and dissipates the heat from the refrigerant. The expansion mechanism 4 is a mechanism for reducing the pressure of the refrigerant, and for example, an expansion valve can be applied.

[0047] In the user-side heat exchanger 5 of the user-side unit, the refrigerant, which has been expanded (depressurized) by the expansion mechanism 4, flows in via the first pipe 6, and heat exchange occurs between this refrigerant and the air in the living space, heating the refrigerant. The user-side unit can lower the temperature of the air discharged from the user-side unit by cooling the air in conjunction with the heat exchange in the user-side heat exchanger 5.

[0048] Furthermore, the air conditioning system 1 may be configured to reverse the direction of refrigerant flow by installing a switching valve (three-way or four-way switching valve) (not shown) between the compressor 2 and the heat source side heat exchanger 3. By switching the flow of the refrigerant in this way, the air conditioning system 1 can perform heating operation in addition to cooling operation.

[0049] The above-described air conditioning system 1 may have its frequency response measured at the site of use, such as during installation or maintenance, for one or both of the controlled object and the control device. This frequency response is a transfer function that includes the characteristics of the amplitude ratio and phase difference between the input signal and output signal at selected points in each component of the controlled object and the control device. For example, in order to properly control the levitation state of the rotating shaft 25, it is necessary to properly measure and recognize the transfer function of the magnetic bearing 20 of the compressor 2. Hereinafter, the work of measuring the frequency response at the site of use will also be referred to as a field test.

[0050] <Calculation System> When conducting field tests on the magnetic bearing 20, the operator constructs a calculation system 100, for example, as shown in Figures 1 and 2. The calculation system 100 is formed by applying a calculation device 50 to a magnetic bearing system consisting of a magnetic bearing control device 10 and a magnetic bearing 20 that are installed at the site of use in conjunction with the installation of the air conditioning system 1. The calculation device 50 acquires input signals and output signals from selected locations in the magnetic bearing control device 10, which is the control device, and the magnetic bearing 20, which is the controlled object, and calculates the frequency response at those locations.

[0051] The arithmetic unit 50 can be a computer having a processor 51, memory 52, and input / output interfaces and communication interfaces (not shown). The processor 51 is a combination of one or more of the following: CPU (Central Processing Unit), GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), and circuits consisting of multiple discrete semiconductors. The memory 52 includes main memory and auxiliary memory. In other words, in this disclosure, the arithmetic unit 50 is an electronic circuit having a CPU, GPU, ASIC, FPGA, etc., and performs various control operations described in this specification by executing instruction codes stored in the memory 52 or by circuit design for special applications. The processor 51 of the arithmetic unit 50 constructs a signal acquisition unit 511 and a calculation unit 512, for example, as shown in Figure 2, by executing a program stored in the memory 52.

[0052] The signal acquisition unit 511 acquires input and output signal data from the components of the magnetic bearing control device 10 and the magnetic bearing 20 selected by the operator, and temporarily stores it in the memory 52. ​​The arithmetic unit 50 may automatically erase the input and output signal data stored in the memory 52 after calculating the frequency response. This allows the arithmetic unit 50 to appropriately secure the capacity of the memory 52 during the process of calculating the frequency response of multiple frequencies.

[0053] Furthermore, the calculation unit 512 is a functional unit that calculates the frequency response based on the input signal and output signal acquired by the signal acquisition unit 511. The calculation of the frequency response by this calculation unit 512 will be described in detail later.

[0054] On the other hand, the magnetic axis control device 10 and the magnetic bearing 20 control the levitation of the rotating shaft 25 of the compressor 2 by various configurations as shown in FIG. 2. Specifically, the magnetic axis control device 10 includes a levitation position command unit 11, a subtractor 12, a position control unit 13, a support force-current conversion unit 14, and a current control unit 15. Each unit of the magnetic axis control device 10 is a software functional unit formed by the processor of the magnetic axis control device 10 executing a program stored in a memory. However, not limited to this, each unit of the magnetic axis control device 10 may be a hardware functional unit formed by an appropriate circuit.

[0055] The levitation position command unit 11 outputs a position command value x of the rotating shaft 25 to the subtractor 12 based on the target position of the rotating shaft 25 received from the control unit of the heat source side unit. For example, the target position of the rotating shaft 25 is defined or calculated at an appropriate position in the control unit according to the type of the compressor 2 and the driving content. ＊ to the subtractor 12.

[0056] The subtractor 12 receives the position command value x output from the levitation position command unit 11 and the actual position x fed back from the gap sensor 28 of the magnetic bearing 20. The subtractor 12 subtracts the position command value x by the actual position x to output the corrected position deviation xEr to the position control unit 13. ＊ ＊

[0057] When the position deviation xEr is input, the position control unit 13 calculates a support force command value f for levitating the rotating shaft 25 by the magnetic bearing 20 and arranging it at an appropriate position based on the position deviation xEr. Then, the position control unit 13 outputs the calculated support force command value f to the support force-current conversion unit 14. ＊ ＊

[0058] When the support force command value f is input to the support force-current conversion unit 14, the support force-current conversion unit 14 calculates a current command value i to be supplied to the magnetic bearing 20 based on the support force command value f and outputs it to the current control unit 15. For example, the support force-current conversion unit 14 calculates the support force command value f ＊ ＊ ＊ ＊ ＊ ​​​​​​​​The system has pre-existing table information that associates the following, and the input support force command value f is entered by referring to the table information. ＊ From current command value i ＊ Extract it.

[0059] The current control unit 15 controls the current command value i ＊ When this is input, this current command value i ＊ The current control unit 15 outputs a control signal for the inverter 21 of the magnetic bearing 20 in accordance with the voltage command value v ＊ Calculate the voltage command value v ＊ Outputs.

[0060] Furthermore, the magnetic bearing 20 of the compressor 2, which is the object of control, comprises an inverter 21, a coil 22, an electromagnet 23, a magnetic bearing rotor 24, and a gap sensor 28. Each component of this magnetic bearing 20 is a hardware functional part formed from appropriate materials.

[0061] The inverter 21 controls the voltage command value v ＊ Upon receiving the signal, the voltage command value v ＊ The coil 22 outputs an actual voltage v corresponding to the frequency. The magnetic bearing 20 is voltage-controlled via the inverter 21 when the actual voltage v corresponding to the frequency of the magnetic bearing 20 is input to the coil 22.

[0062] The coil 22 is supplied with power at the actual voltage v output from the inverter 21. This allows the coil 22 to supply a controlled actual current i to the electromagnets 23. Each electromagnet 23 is excited based on the actual current i of the coil 22, generating an actual support force f that levitates the rotating shaft 25. The levitation position of the rotating shaft 25 is adjusted based on the actual support force f (magnetic field) of each electromagnet 23 orbiting the rotating shaft 25.

[0063] The gap sensor 28 detects the gap dimension of the rotating shaft 25 relative to the magnetic bearing 20 when the rotating shaft 25 is levitating, and transmits the detected information (actual position x of the rotating shaft 25) to the magnetic bearing control device 10. The magnetic bearing control device 10 feeds back this actual position of the rotating shaft 25 to the subtractor 12. The subtractor 12 then processes the position command value x of the rotating shaft 25 as described above. ＊The position of the rotation axis 25 is corrected by subtracting the actual position x from this value.

[0064] The operator then constructs a calculation system 100 by connecting the calculation device 50 to the points in the configuration of the magnetic bearing control device 10 and the magnetic bearing 20 where frequency response is to be obtained. The calculation system 100 outputs an excitation signal with a set and controlled frequency from the magnetic bearing control device 10 to the magnetic bearing 20, and the calculation device 50 acquires the input signal input to a selected point in the control loop into which the excitation signal is input, and the output signal output from the selected point in the control loop into which the excitation signal is input. The excitation signal may be supplied from outside the control loop as shown in Figure 10, but the input signal and output signal for calculating the frequency response are acquired from within the control loop. In other words, the "input signal" is the signal input to the point in the control loop of the magnetic bearing control device 10 and the magnetic bearing 20 where frequency response is to be obtained. This input signal may be the excitation signal initially input from the signal generation unit 30, or it may be a signal (command value, actual voltage, actual current, actual position) input to each part of the magnetic bearing control device 10 and the magnetic bearing 20. Furthermore, the "output signal" is the signal output from the point in the control loop of the magnetic bearing control device 10 and the magnetic bearing 20 from which the frequency response is to be obtained. By acquiring and processing this input signal and output signal, the arithmetic unit 50 can calculate the frequency response of the selected point.

[0065] The part of the magnetic bearing control device 10 and magnetic bearing 20 selected by the operator may be all of the components of the magnetic bearing control device 10 and magnetic bearing 20, or only a part of them. For example, when measuring the entire magnetic bearing control device 10 and magnetic bearing 20, the position command value x input to the levitation position command unit 11 of the magnetic bearing control device 10 is used. ＊ The frequency response between the input signal and the actual position x of the gap sensor 28 of the magnetic bearing 20 (output signal) is calculated. For example, when measuring the frequency response of the current control system of the magnetic bearing control device 10 and the magnetic bearing 20, the current command value i is used as the input signal. ＊The frequency response between the actual current i and the output signal is calculated. Alternatively, the calculation system 100 may calculate the frequency response only within the magnetic bearing control device 10. As an example, when checking the frequency response of the position control unit 13, the position deviation xEr (input signal) input to the position control unit 13 and the support force command value f output from the position control unit 13 are used. ＊ Calculate the frequency response of the output signal.

[0066] The calculation system 100 described above includes a signal generation unit 30 and a user setting unit 31 that allows the user to set the state of the excitation signal in order to output an excitation signal from the magnetic bearing control device 10 to the magnetic bearing 20 in the measurement of the frequency response. Figure 2 illustrates a configuration in which the signal generation unit 30 and the user setting unit 31 are pre-installed inside the magnetic bearing control device 10.

[0067] The user setting unit 31, for example, works in conjunction with the operation unit and display unit of the magnetic bearing control device 10 to allow the operator to set the state of the excitation signal to be recognized. Examples of the excitation signal state that the operator can set include the frequency band, the number of frequencies (the resolution of the frequencies used to calculate the frequency response), and the amplitude. The user setting unit 31 may be configured to allow the operator to set all of the excitation signal states, or it may be configured to allow the operator to set only some of them.

[0068] The signal generation unit 30 generates an excitation signal based on the settings in the user setting unit 31 and outputs the excitation signal to the levitation position command unit 11 of the magnetic bearing control device 10. The "excitation signal" is a signal that has a changing frequency to confirm the operation of the magnetic bearing 20, is output from the signal generation unit 30 and applied to the magnetic bearing 20 which is the control target, and causes the magnetic bearing 20 to vibrate. In this embodiment, the signal generation unit 30 generates a divided multisine signal as the excitation signal, which has a first waveform formed by superimposing multiple sine waves of different frequencies and a second waveform formed by superimposing multiple sine waves of different frequencies, but with a different waveform from the first waveform, on different time axes.

[0069] <Divided Multisine Signal> The significance of using this divided multisine signal and the types of divided multisine signals will be explained below with reference to Figures 3 to 6. When measuring the frequency response of a control device and a controlled object, multiple excitation signals of different frequencies are usually generated and output sequentially, and the frequency response for each of the multiple frequencies is measured. For example, the frequency response is calculated for each frequency while increasing the frequency from a low frequency to a high frequency within the frequency band of the excitation signal. In this case, the frequency of the single waveform of the excitation signal will change at appropriate intervals. In particular, when measuring the frequency response of a large number of frequencies (for example, several hundred to several thousand frequencies), it will take a considerable amount of time.

[0070] Therefore, in calculating the frequency response, a method is known in which a multisine signal is generated by superimposing multiple sine waves of different frequencies, as shown in Figure 3, and output as an excitation signal, and the frequency responses of multiple frequencies contained in the multisine signal are calculated simultaneously. For the sake of ease of understanding, Figure 3 illustrates a multisine signal obtained by adding up six sine waves of different frequencies. Each sine wave contained in the multisine signal can be decomposed into separate sine waves when calculating the frequency response, and the frequency response of each sine wave can be calculated.

[0071] However, the more the frequency band to be measured or the number of superimposed sine waves increases, the greater the processing load on the computing device and the memory capacity of the multisine signal. In field tests at the site where the above-described air conditioning system 1 is installed, it may be difficult to use a high-performance computing device for analysis, for example, because it cannot be brought in or because it takes time to arrange and transport it. Furthermore, in field tests, inspections and maintenance of other parts of the air conditioning system 1 are also performed, so there is often not much time available. Therefore, in the calculation system 100 according to the embodiment, a divided multisine signal is generated by changing the sine waves contained in the excitation signal on the time axis of the excitation signal, according to the performance of the computing device 50.

[0072] As an example of the divided multisine signal generated by the signal generation unit 30, the first example shown in Figure 4 can be applied. The divided multisine signal of the first example is a pattern in which multiple sine waves included in the divided multisine signal are switched at the same timing. That is, the divided multisine signal of the first example changes all the superimposed sine waves simultaneously at each set time point, thereby making the waveform different for each range set on the time axis.

[0073] In Figure 4, the divided multisine signal has a first waveform during the period from time t1 to time t2, a second waveform during the period from time t2 to time t3, and a third waveform during the period from time t3 to time t4. The first waveform is formed by superimposing a sine wave of frequency f11, a sine wave of frequency f12, a sine wave of frequency f13, ... a sine wave of frequency f1n. Similarly, the second waveform is formed by superimposing a sine wave of frequency f21, a sine wave of frequency f22, a sine wave of frequency f23, ... a sine wave of frequency f2n. The third waveform is formed by superimposing a sine wave of frequency f31, a sine wave of frequency f32, a sine wave of frequency f33, ... a sine wave of frequency f3n.

[0074] In this case, the periods of the first, second, third, ... nth waveforms may be the same length or may be different lengths. If the periods are the same length, the signal generation unit 30 can easily manage each waveform and the sine waves contained in them. On the other hand, if the periods are of different lengths, the signal generation unit 30 can shorten the period as the magnitude of the frequency contained in each waveform increases, for example. In other words, the periods of the first, second, third, ... nth waveforms may be set according to the sine wave with the smallest frequency among the sine waves contained in each waveform. This makes it possible to shorten the overall duration of the field test.

[0075] When generating a divided multisine signal, it is preferable to generate each waveform (first waveform, second waveform, third waveform, ... nth waveform) by superimposing sine waves of similar frequencies. For example, when the frequency band to be measured is 1 Hz to 10 Hz and the number of frequencies is 10 (in 1 Hz increments), when forming the divided multisine signal of the first and second waveforms, a first waveform is generated by superimposing a total of five sine waves from 1 Hz to 5 Hz, and a second waveform is generated by superimposing a total of five sine waves from 6 Hz to 10 Hz.

[0076] When calculating the frequency response for each sine wave of multiple frequencies, at least one period of waveform is required for each frequency sine wave. Lower frequency sine waves have longer periods. Therefore, if a 10 Hz sine wave and a 1 Hz sine wave are combined, the 10 Hz sine wave will have to be repeated multiple times until the period of the 1 Hz sine wave is completed, which will take a long time. In contrast, by superimposing sine waves of similar frequencies, it is possible to shorten the duration of each waveform in the divided multisine signal.

[0077] Furthermore, while the sine waves 1, 2, 3, ..., n in Figure 4 are set to have different amplitudes from each other, the first and second waveforms, which have different time axes, may have the same amplitude. For example, the sine waves with frequencies f11, f21, f31, ..., fn1 in sine wave 1 can have the same amplitude. The same applies to sine waves 2, 3, ..., n. However, this is not limited to this; for example, the sine waves with frequencies f11, f21, f31, ..., fn1 in sine wave 1 may have partially or entirely different amplitudes. The same applies to sine waves 2, 3, ..., n.

[0078] Furthermore, it is preferable to set the amplitude of each sine wave forming each waveform to be larger for sine waves with higher frequencies. This is because the inertial force of the magnetic bearing 20 increases as the frequency increases, and the ratio of the output signal to the excitation signal (gain) decreases. For example, if the relationship of the frequencies of each sine wave in the first waveform is frequency f11 > frequency f12 > frequency f13 > ... frequency f1n, then it is preferable that the amplitude also follows the relationship frequency f11 > frequency f12 > frequency f13 > ... frequency f1n.

[0079] Alternatively, the signal generation unit 30 may apply the second example shown in Figure 5 as the divided multisine signal to be generated. The divided multisine signal of the second example is a pattern in which multiple sine waves included in the divided multisine signal are switched at different timings. That is, the divided multisine signal of the second example switches one (or more) sine waves of any frequency while maintaining sine waves of other frequencies, thereby switching sequentially to the first waveform, second waveform, third waveform, ... nth waveform.

[0080] In Figure 5, the divided multisine signal is formed in the first waveform from time t1 to time t2 by superimposing sine waves of frequency f11, frequency f12, frequency f13, ... frequency f1n. Then, at the switch at time t2, only the sine wave of frequency f11 switches to the sine wave of frequency f12, while the sine waves of the other frequencies do not switch. In other words, the second waveform from time t2 to time t3 is formed by superimposing sine waves of frequency f21, frequency f12, frequency f13, ... frequency f1n. Similarly, at the switch at time t3, only the sine wave of frequency f12 switches to the sine wave of frequency f22, while the sine waves of the other frequencies do not switch. In other words, the third waveform at time points t3 to t4 is formed by superimposing a sine wave of frequency f21, a sine wave of frequency f22, a sine wave of frequency f13, ..., a sine wave of frequency f1n. Similarly, at the subsequent transitions between individual sine waves, the waveform of the divided multisine signal changes to a different form.

[0081] Furthermore, the split multisine signal in the second example is not limited to the switching of a single sine wave at the same time; multiple sine waves of different frequencies may switch simultaneously. For example, Figure 5 illustrates an example where, at time t5, the sine wave of frequency f21 and the sine wave of frequency f1n each switch simultaneously to the sine wave of frequency f31 and the sine wave of frequency f2n, respectively.

[0082] The switching timing for each sine wave (in other words, the period during which each frequency sine wave is output) should be determined according to the frequency of each sine wave. For example, the higher the frequency of the sine wave, the earlier the switching timing can be set. This is because a higher frequency sine wave has a greater number of repetitions of amplitude (period) within the set period.

[0083] Furthermore, while the sine waves 1, 2, 3, ..., n in Figure 5 are set to have different amplitudes from each other, the amplitudes of the sine waves that switch at each point in time may be the same or different. For example, by setting the sine waves with frequencies f11, f21, f31, ..., fn1 in sine wave 1 to have the same amplitude, and then setting the sine waves 2, 3, ..., n to have the same amplitude, the management of each sine wave can be simplified.

[0084] Furthermore, in the first example shown in Figure 4 and the second example shown in Figure 5, the signal generation unit 30 sets the number of sine waves forming each waveform of the divided multisine signal to be the same on the time axis. By setting the number of sine waves in each waveform to be the same in this way, it becomes possible to output the excitation signal efficiently and shorten the time required to calculate the frequency response.

[0085] However, as shown in the third example in Figure 6, the divided multisine signal may have the frequency and amplitude of one or more of the multiple sine waves forming each waveform set to zero (a configuration in which no sine waves are output). For example, Figure 6 shows an example in the divided multisine signal of the second example in which some of the multiple sine waves are set to zero. As an example, in the divided multisine signal of the second example, the waveform switches in accordance with the switching of one sine wave, but in some cases it may be better to switch multiple sine waves simultaneously. One example of this is when resonance occurs depending on the combination of frequencies. In such cases, the signal generation unit 30 can set some of the sine waves to zero and switch multiple sine waves at an arbitrary point in time in sync.

[0086] In Figure 6, some sine waves are set to zero in the second example of the divided multisine signal, but some sine waves may also be set to zero in the first example of the divided multisine signal. For example, in the initial stage when the arithmetic unit 50 calculates the frequency response, the waveform of the divided multisine signal may be formed using a large number of sine waves, while in the later stage when the arithmetic unit 50 calculates the frequency response, each waveform of the divided multisine signal may be formed using a small number of sine waves. This is because in the later stage when the frequency response is calculated, there is a possibility that the processing capacity of the arithmetic unit 50 may decrease due to an increase in the cache memory, etc.

[0087] The signal generation unit 30 may automatically set the number and duration of sine waves included in the waveform of the divided multisine signal, the number of divisions and duration of each waveform, etc., based on the frequency bandwidth and number of frequencies (frequency resolution for calculating the frequency response) set by the operator and the performance of the arithmetic unit 50. Examples of the performance of the arithmetic unit 50 include the number of calculations per unit time of the processor 51 and the capacity of the memory 52 that stores the input and output signals.

[0088] For example, if the number of operations processed per unit time by the processor 51 is large, the number of sine waves included in the waveform of the divided multisine signal is increased, thereby reducing the number of divisions in each waveform of the divided multisine signal. Conversely, if the number of operations processed per unit time by the processor 51 is small, the number of sine waves included in the waveform of the divided multisine signal is decreased, thereby increasing the number of divisions in each waveform of the divided multisine signal. Alternatively, if the capacity of the memory 52 is large, the number of sine waves included in the waveform of the divided multisine signal is increased, thereby reducing the number of divisions in each waveform of the divided multisine signal. Conversely, if the capacity of the memory 52 is small, the number of sine waves included in the waveform of the divided multisine signal is decreased, thereby increasing the number of divisions in each waveform of the divided multisine signal.

[0089] However, the actual performance of the arithmetic unit 50 is the sum of the number of operations processed per unit time by the processor 51 and the capacity of the memory 52. ​​For this reason, the signal generation unit 30 should combine this information to set the number of sine waves included in each waveform of the divided multisine signal. Of course, the performance of the arithmetic unit 50 also varies depending on other factors (thermal control, communication speed, etc.), so each waveform of the divided multisine signal may be generated based on various factors.

[0090] <Calculation Method> The air conditioning device 1 and calculation system 100 according to this embodiment are basically configured as described above, and their operation will be explained below with reference to the flowchart in Figure 7. The magnetic bearing control device 10 and the calculation device 50 of the calculation system 100 cooperate with each other to perform steps S101 to S107 in Figure 7 as a method for calculating the frequency response.

[0091] In the calculation method, the operator first selects the locations in the magnetic bearing control device 10 and the magnetic bearing 20 where the frequency response is to be obtained, in other words, the locations of the input and output signals to be acquired by the calculation device 50 (step S101). As described above, the calculation system 100 can selectively acquire the frequency response for some or all of the configurations of the magnetic bearing control device 10, which is the control device, and the magnetic bearing 20, which is the controlled object.

[0092] Next, in the calculation method, the operator sets the frequency band and the number of frequencies (resolution), which are the range for calculating the frequency response, via the user setting unit 31 (step S102). For example, the frequency band of the magnetic bearing 20 may be in the range of 1 Hz to 2000 Hz. The frequency resolution of the magnetic bearing 20 may be in units of 1 Hz (in the case of the range of 1 Hz to 2000 Hz, the number of frequencies is 2000).

[0093] Then, the signal generation unit 30 sets the number of sine waves included in each waveform of the divided multisine signal and the number of divisions of each waveform based on the set frequency band and number of frequencies and the performance of the arithmetic unit 50 (step S103). As described above, if the performance of the arithmetic unit 50 is high, it is good to set a large number of sine waves included in each waveform of the divided multisine signal. On the other hand, if the performance of the arithmetic unit 50 is low, it is good to set a small number of sine waves included in each waveform of the divided multisine signal.

[0094] Once step S103 is completed, the calculation system 100 moves to the stage of actually acquiring the frequency response in a field test. In this case, the signal generation unit 30 generates each waveform of the divided multisine signal along the time axis according to the settings of the above steps, and outputs the divided multisine signal to the magnetic bearing 20, which is the target of control, via the magnetic bearing control device 10 (step S104).

[0095] The signal acquisition unit 511 of the arithmetic unit 50 acquires the input signal and output signal when each waveform of the divided multisine signal is input at a location selected by the operator from among the components of the magnetic bearing control device 10 and the magnetic bearing 20 (step S105).

[0096] Then, the calculation unit 512 of the arithmetic unit 50 calculates the frequency response by discrete Fourier transform while acquiring the input signal and output signal by the signal acquisition unit 511 (step S106). Specifically, the calculation unit 512 acquires the input signal as shown in the upper left figure of Figure 8, and the output signal as shown in the upper right figure of Figure 8. The calculation unit 512 performs a discrete Fourier transform (Fast Fourier transform) DFT process on these input and output signals (hereinafter also referred to as input and output signals) and calculates the frequency response for each sine wave contained in the waveform. The frequency response for each of the multiple (1 Hz, 2 Hz, ...) sine waves is calculated including the amplitude ratio and phase difference characteristics of the input and output signals for each frequency, as shown in the lower figure of Figure 8, (1), (2), and (3).

[0097] Specifically, the calculation unit 512 performs the operations shown in Figures 9(A) and 9(B) during the DFT process. In the DFT process, as shown in Figure 9(A), for each of the N intervals in each waveform (time-series waveform) of the divided multisine signal, the discrete signals x(n) (x(0), x(1), x(2), ..., x(N-1)) are added together at each time step using the following formula (1) of the discrete Fourier transform.

[0098]

[0099] The calculation unit 512 can calculate the frequency responses X(0), X(1), X(2), ... X(N-1) for each sine wave contained in each waveform by summing N values ​​calculated using formula (1) with discrete signals x(0), x(1), x(2), ... X(N-1), as shown in Figure 9(B). In this case, the frequency responses X(0), X(1), X(2), ... X(N-1) for each sine wave are independent and do not influence each other. Therefore, the calculation unit 512 can calculate the frequency responses X(0), X(1), X(2), ... X(N-1) in parallel. The frequency responses X(0), X(1), X(2), ... X(N-1) for each sine wave include amplitude ratio and phase difference characteristics, as shown in the left figure of Figure 9(B). In this way, the calculation unit 512 can accurately calculate the frequency responses for multiple sine waves with different frequencies.

[0100] For example, the calculation unit 512 calculates the frequency response (amplitude ratio or phase difference) for each of the multiple sine waves forming the first waveform by adding up the values ​​calculated using formula (1) with the discrete signals calculated for each arbitrary time point of the first waveform. Once the calculation of the first waveform is complete, the calculation unit 512 erases the discrete signals used in the calculation. Then, for example, the calculation unit 512 calculates the frequency response (amplitude ratio or phase difference) for each of the multiple sine waves forming the second waveform by adding up the values ​​calculated using formula (1) with the discrete signals calculated for each arbitrary time point of the second waveform. The calculation unit 512 can obtain the frequency response of all the multiple sine waves included in all the waveforms by repeating this operation for each waveform of the divided multisine signal.

[0101] Returning to Figure 7, in step S107 following step S106, the arithmetic unit 50 determines whether or not the frequency response has been measured for all waveforms (sine waves of all frequencies) of the divided multisine signal. If the frequency response of some of the waveforms of the divided multisine signal has not been measured (step S107: NO), the process returns to step S104 and the same processing flow is repeated. On the other hand, if the frequency response of all waveforms of the divided multisine signal has been measured (step S107: YES), the processing flow of the calculation method is terminated.

[0102] As described above, the computing unit 50 can accurately calculate the frequency response of selected locations in each component of the magnetic bearing control device 10 and the magnetic bearing 20 for multiple frequencies. By knowing this frequency response, the operator can understand the stability of signal transmission, or delays or abnormalities in signal transmission, in accordance with the frequency in the magnetic bearing control device 10 and the magnetic bearing 20. For example, the operator can use the understood frequency response to identify maintenance targets for the magnetic bearing control device 10 and the magnetic bearing 20 based on the frequency response. Furthermore, in controlling the magnetic bearing 20, the magnetic bearing system can automatically or manually correct the software of the magnetic bearing control device 10 to adjust the power supply state to the magnetic bearing 20 in accordance with the frequency response.

[0103] Furthermore, the calculation system 100 and the refrigeration apparatus 1 according to this disclosure are not limited to the above-described embodiment and can take various modified forms. For example, in the above embodiment, the frequency response is calculated by a separate computer, the arithmetic unit 50, from the magnetic bearing control device 10. However, the calculation system 100 may consist only of the magnetic bearing control device 10 and the magnetic bearing 20, by incorporating the functions of the above-described arithmetic unit 50 (signal acquisition unit 511, calculation unit 512) within the magnetic bearing control device 10. In addition, the calculation system 100 is not limited to application to the refrigeration apparatus 1 or the magnetic bearing system, but may be applied to various devices that require the calculation of frequency response. Moreover, the calculation system 100 is not limited to a configuration that calculates the frequency response of a device installed at a site of use, but may also be applied to devices in research institutes, testing facilities, factories, and other places where devices are developed, manufactured, etc.

[0104] Furthermore, as shown in the modified example in Figure 10, the signal generation unit 30 and the user setting unit 31 are not limited to being provided inside the magnetic bearing control device 10, but may be provided outside the magnetic bearing control device 10. For example, Figure 10 shows an example in which an oscillator 32 having a signal generation unit 30 and a user setting unit 31 is connected to the magnetic bearing control device 10, and a divided multisine signal is output from the oscillator 32. This allows the calculation system 100 to simplify the configuration of the magnetic bearing control device 10, thereby promoting miniaturization of the device, cost reduction, etc. Alternatively, the calculation system 100 may be configured to include a signal generation unit 30 and a user setting unit 31 in the arithmetic unit 50, and connect the arithmetic unit 50 to the magnetic bearing control device 10 to perform both the output of the excitation signal and the calculation of the frequency response.

[0105] <Modes and Effects of the Disclosure> The embodiments disclosed above have, for example, the following modes and effects.

[0106] [Note 1] A calculation system comprising: a signal generation unit that inputs an excitation signal from a control device to a controlled object for the purpose of exciting the controlled object; and a calculation unit that calculates the frequency response of part or all of the control device and the controlled object when the excitation signal is input to the controlled object, wherein the excitation signal is a divided multisine signal having a first waveform obtained by superimposing a plurality of sine waves of different frequencies and a second waveform obtained by superimposing a plurality of sine waves of different frequencies and having a waveform different from the first waveform, on different time axes.

[0107] [Effects of Appendix 1] According to the above, when the calculation system calculates the frequency response, it uses a divided multisine signal as the excitation signal, allowing it to input multiple waveforms with an appropriate number of sine waves to the controlled object by shifting them on the time axis according to the performance of the computing device. This makes it possible to efficiently obtain the frequency response of the control device and the controlled object according to the performance of the computing device. For example, the calculation system can obtain good frequency characteristics of the controlled object even in usage sites where it is difficult to apply a high-performance computing device.

[0108] [Note 2] The calculation system described in Note 1, wherein the signal generation unit switches all of the plurality of sine waves forming the first waveform and all of the plurality of sine waves forming the second waveform at the same timing.

[0109] [Effect of Appendix 2] In this way, by switching all of the multiple sine waves of the first waveform and all of the multiple sine waves of the second waveform at the same timing, the calculation system can easily manage the multiple sine waves used to calculate the frequency response.

[0110] [Note 3] The calculation system described in Note 1, wherein the signal generation unit switches from the first waveform to the second waveform by switching some of the multiple sine waves forming the first waveform to sine waves of different frequencies.

[0111] [Effect of Appendix 3] In this way, by switching some of the sine waves among the multiple sine waves forming the first waveform to sine waves of different frequencies, the calculation system can calculate the frequency response of all sine waves more efficiently.

[0112] [Appendix 4] The calculation system according to any one of Appendix 1 to 3, wherein the number of sine waves forming the first waveform is the same as the number of sine waves forming the second waveform.

[0113] [Effect of Appendix 4] Thus, by having the same number of sine waves forming the first waveform as the same number of sine waves forming the second waveform, the calculation system can calculate the frequency response of the multiple sine waves more efficiently.

[0114] [Note 5] The calculation system according to any one of Notes 1 to 4, wherein the signal generation unit sets the amplitude of the multiple sine waves of the divided multisine signal to be larger for sine waves with higher frequencies.

[0115] [Effect of Appendix 5] As a result, the calculation system can accurately calculate the frequency response even for high-frequency sine waves.

[0116] [Note 6] The calculation system according to any one of Notes 1 to 5, wherein the frequency response includes the characteristics of the amplitude ratio and phase difference between the input signal and the output signal, and the calculation unit calculates the amplitude ratio or the phase difference for each of the multiple sine waves by performing a discrete Fourier transform on the input signal and the output signal acquired from the control device or the controlled object.

[0117] [Effects of Appendix 6] In this way, by performing the Discrete Fourier Transform, the calculation system can easily calculate the frequency response for each sine wave of multiple frequencies.

[0118] [Note 7] The calculation unit calculates the amplitude ratio or phase difference for each of the multiple sine waves forming the first waveform by adding up the values ​​calculated for each arbitrary time point of the first waveform, and calculates the amplitude ratio or phase difference for each of the multiple sine waves forming the second waveform by adding up the values ​​calculated for each arbitrary time point of the second waveform, as described in Note 6.

[0119] [Effects of Appendix 7] In this way, the discrete Fourier transform allows for the calculation of the amplitude ratio or phase difference of the frequency response independently of each other by summing the values ​​calculated at any given time in the first and second waveforms. As a result, the data used can be immediately erased, and the increase in the memory capacity of the arithmetic unit can be suppressed.

[0120] [Note 8] The signal generation unit is a calculation system according to any one of Notes 1 to 7, provided inside the control device.

[0121] [Effects of Appendix 8] By providing the signal generation unit inside the control device in this way, the hassle of preparing a separate signal generation unit is eliminated, and the frequency response can be obtained more easily.

[0122] [Note 9] The signal generation unit is a calculation system according to any one of Notes 1 to 7, provided outside the control device.

[0123] [Effects of Appendix 9] By providing the signal generation unit outside the control device in this way, it is possible to suppress the increase in size of the control device.

[0124] [Note 10] The calculation system according to any one of Notes 1 to 9, comprising a user setting unit that allows the user to set the frequency band or the number of frequencies of the excitation signal input to the controlled object, wherein the signal generation unit automatically sets the number of multiple sine waves included in the first waveform and the second waveform of the divided multisine signal based on the frequency band or the number of frequencies set by the user setting unit.

[0125] [Effects of Appendix 10] As a result, the calculation system can suppress the load set by the user when acquiring the frequency response, and can also appropriately generate a divided multisine signal.

[0126] [Appendix 11] A magnetic bearing system comprising: a calculation system described in any one of Appendix 1 to 10; a magnetic bearing that non-contactually supports a rotating shaft; and a magnetic bearing control device for controlling the magnetic bearing, wherein the controlled object is a magnetic bearing, and the control device is a magnetic bearing control device.

[0127] [Effects of Appendix 11] As a result, the calculation system can stably measure the frequency response of the magnetic bearing system.

[0128] [Note 12] A refrigeration apparatus comprising: a compressor having a magnetic bearing as a control object; a control device for controlling the operation of the magnetic bearing; and a refrigerant circuit provided with the compressor for circulating a refrigerant based on the operation of the compressor, wherein the control device inputs an excitation signal to the magnetic bearing for excitation of the magnetic bearing, and causes a calculation unit to calculate the frequency response of part or all of the magnetic bearing and the control device when the excitation signal is input to the magnetic bearing, and the excitation signal is a divided multisine signal having a first waveform obtained by superimposing a plurality of sine waves of different frequencies and a second waveform obtained by superimposing a plurality of sine waves of different frequencies and having a waveform different from the first waveform, on different time axes.

[0129] [Effects of Appendix 12] Even in this case, the refrigeration system can obtain good frequency characteristics of the magnetic bearings, even at the site where the compressor is in use.

[0130] The refrigeration system (air conditioning system 1), calculation system 100, and magnetic bearing system according to the embodiments disclosed herein are illustrative and not restrictive in all respects. The embodiments can be modified and improved in various ways without departing from the scope and spirit of the appended claims. The matters described in the above embodiments can be otherwise configured and combined in a non-consistent manner.

[0131] This application claims priority to Japanese Patent Application No. 2024-168835, a basic application filed with the Japan Patent Office on September 27, 2024, the entire contents of which are incorporated herein by reference.

[0132] 1 Refrigeration system (air conditioning system) 2 Compressor 8 Refrigerant circuit 10 Magnetic bearing control device 20 Magnetic bearing 25 Rotating shaft 30 Signal generation unit 31 User setting unit 50 Calculation unit 512 Calculation unit

Claims

1. A calculation system (100) comprising: a signal generation unit (30) that causes a control device (10) to input an excitation signal to a controlled object (20) for the purpose of exciting the controlled object (20); and a calculation unit (512) that calculates the frequency response of part or all of the control device (10) and the controlled object (20) when the excitation signal is input to the controlled object (20), wherein the excitation signal is a divided multisine signal having a first waveform obtained by superimposing a plurality of sine waves of different frequencies and a second waveform obtained by superimposing a plurality of sine waves of different frequencies and having a waveform different from the first waveform, on different time axes.

2. The calculation system (100) according to claim 1, wherein the signal generation unit (30) switches all of the plurality of sine waves forming the first waveform and all of the plurality of sine waves forming the second waveform at the same timing.

3. The calculation system (100) according to claim 1, wherein the signal generation unit (30) switches from the first waveform to the second waveform by switching some of the plurality of sine waves forming the first waveform to sine waves of different frequencies.

4. The calculation system (100) according to any one of claims 1 to 3, wherein the number of sine waves forming the first waveform is the same as the number of sine waves forming the second waveform.

5. The calculation system (100) according to any one of claims 1 to 4, wherein the signal generation unit (30) sets the amplitude of the multiple sine waves of the divided multisine signal to be larger for sine waves with higher frequencies.

6. The calculation system (100) according to any one of claims 1 to 5, wherein the frequency response includes the characteristics of the amplitude ratio and phase difference between the input signal and the output signal, and the calculation unit (512) calculates the amplitude ratio or the phase difference for each of the plurality of sine waves by performing a discrete Fourier transform on the input signal and the output signal acquired from the control device (10) or the controlled object (20).

7. The calculation unit (512) calculates the amplitude ratio or phase difference for each of the plurality of sine waves forming the first waveform by summing the values ​​calculated for each arbitrary time of the first waveform, and calculates the amplitude ratio or phase difference for each of the plurality of sine waves forming the second waveform by summing the values ​​calculated for each arbitrary time of the second waveform, the calculation system (100) according to claim 6.

8. The calculation system (100) according to any one of claims 1 to 7, wherein the signal generation unit (30) is provided inside the control device (10).

9. The calculation system (100) according to any one of claims 1 to 7, wherein the signal generation unit (30) is provided outside the control device (10).

10. A calculation system (100) according to any one of claims 1 to 9, comprising a user setting unit (31) that allows the user to set the frequency band or the number of frequencies of the excitation signal to be input to the controlled object (20), wherein the signal generation unit (30) automatically sets the number of multiple sine waves included in the first waveform and the second waveform of the divided multisine signal based on the frequency band or the number of frequencies set by the user setting unit (31).

11. A magnetic bearing system comprising: a calculation system (100) according to any one of claims 1 to 10; a magnetic bearing (20) that non-contactually supports a rotating shaft (25); and a magnetic bearing control device (10) that controls the magnetic bearing (20), wherein the controlled object (20) is a magnetic bearing, and the control device (10) is a magnetic bearing control device.

12. A refrigeration system (1) comprising: a compressor (2) having a magnetic bearing (20) as a control object; a control device (10) for controlling the operation of the magnetic bearing (20); and a refrigerant circuit (8) on which the compressor (2) is provided and which circulates a refrigerant based on the operation of the compressor (2), wherein the control device (10) inputs an excitation signal to the magnetic bearing (20) for causing the magnetic bearing (20) to vibrate, and causes a calculation unit (512) to calculate the frequency response of part or all of the magnetic bearing (20) and the control device (10) when the excitation signal is input to the magnetic bearing (20), and the excitation signal is a divided multisine signal having a first waveform obtained by superimposing a plurality of sine waves of different frequencies and a second waveform obtained by superimposing a plurality of sine waves of different frequencies and having a waveform different from the first waveform, on different time axes.

Images