Control device, electric device, control method, control program, and control system

Abstract

A control device (11) comprises: a data acquisition unit (113) that acquires first input data which is correlated with first lifespan data relating to the lifespan of a first electric motor (12A); an estimation unit (112) that uses an estimation model (131), which is trained to estimate first lifespan data on the basis of first input data, to estimate first lifespan data on the basis of the first input data acquired by the data acquisition unit (113); and a training unit (111) that retrains the estimation model (131) on the basis of second input data correlated with second lifespan data relating to the lifespan of a second electric motor (12B, 12C).

Description

Control device, electrical device, control method, control program, and control system

[0001] The present disclosure relates to a control device for controlling an electric motor, an electric device, a control method, a control program, and a control system.

[0002] Various electrical appliances, such as air conditioners and vehicles, are equipped with electric motors. If users could know the lifespan of an electric motor in advance, they could replace or repair the motor before the end of its lifespan. In this regard, techniques for estimating the lifespan of an electric motor in advance are known. For example, Japanese Patent Laid-Open Publication No. 2017-46540 (Patent Document 1) discloses a machine learning device that estimates the lifespan of an electric motor based on at least one of output data from a sensor that detects the operating state of the electric motor and data regarding the presence or absence of a malfunction of the electric motor.

[0003] JP 2017-46540 A

[0004] The machine learning device disclosed in JP 2017-46540 A can estimate the lifespan of an electric motor according to the usage environment, and if the accuracy of the lifespan estimation of an electric motor is improved, users will not misjudge the lifespan of the electric motor and will be able to maintain the electric motor appropriately.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a technique that can estimate the life of an electric motor with high accuracy.

[0006] The control device according to the present disclosure includes a data acquisition unit that acquires first input data that is correlated with first life data related to the life of a first electric motor, an estimation unit that estimates the first life data based on the first input data acquired by the data acquisition unit using an estimation model trained to estimate the first life data based on the first input data, and a training unit that retrains the estimation model based on second input data that is correlated with second life data related to the life of a second electric motor.

[0007] The control method according to the present disclosure includes, as processing executed by a computer, a step of acquiring first input data correlated with first life data relating to the life of a first electric motor, a step of estimating the first life data based on the first input data acquired by the acquiring step using an estimation model trained to estimate the first life data based on the first input data, and a step of retraining the estimation model based on second input data correlated with second life data relating to the life of a second electric motor.

[0008] The control program according to the present disclosure causes a computer to execute the steps of acquiring first input data correlated with first life data relating to the life of a first electric motor, estimating the first life data based on the first input data acquired by the acquiring step using an estimation model trained to estimate the first life data based on the first input data, and retraining the estimation model based on second input data correlated with second life data relating to the life of a second electric motor.

[0009] A control system according to the present disclosure includes a plurality of electric devices including a first electric device and a second electric device, and a server device communicatively connected to the plurality of electric devices. The server device includes a data acquisition unit that acquires first input data correlated with first life data relating to a life of a first electric motor included in the first electric device, an estimation unit that estimates the first life data based on the first input data acquired by the data acquisition unit using an estimation model trained to estimate the first life data based on the first input data, and a training unit that retrains the estimation model based on second input data correlated with second life data relating to a life of a second electric motor included in the second electric device.

[0010] According to the present disclosure, an estimation model trained to estimate first lifespan data based on first input data correlated with first lifespan data related to the lifespan of a first electric motor is retrained based on second input data correlated with second lifespan data related to the lifespan of a second electric motor, thereby improving the estimation accuracy of the estimation model by taking into account the lifespan of not only the first electric motor but also the second electric motor. This allows a user to estimate the lifespan of an electric motor with high accuracy using the estimation model.

[0011] FIG. 1 is a diagram showing an example of the configuration of a control system according to a first embodiment. FIG. 2 is a diagram showing an example of the configuration of an electric motor included in an electrical device according to the first embodiment. FIG. 3 is a diagram showing an example of the configuration of an electric motor interface included in a control device according to the first embodiment. FIG. 4 is a diagram showing an example of the configuration of a drive circuit included in a control device according to the first embodiment. FIG. 5 is a diagram showing the functional configuration of a control device according to the first embodiment. FIG. 6 is a diagram for explaining an overview of supervised learning. FIG. 7 is a diagram for explaining input and output of supervised learning in a control device according to the first embodiment. FIG. 8 is a diagram showing an overview of training processing performed by a control device according to the first embodiment. FIG. 9 is a diagram showing the configuration of a neural network. FIG. 10 is a diagram showing an overview of estimation processing performed by a control device according to the first embodiment. FIG. 11 is a flowchart relating to processing performed in a training phase by a control device according to the first embodiment. FIG. 12 is a flowchart relating to processing performed in a utilization phase by a control device according to the first embodiment. FIG. 13 is a diagram showing the configuration of a control system according to a second embodiment. FIG. 14 is a diagram showing the functional configuration of a server device (control device) according to the second embodiment.

[0012] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. While multiple embodiments will be described below, it is anticipated from the beginning that the configurations described in each embodiment will be appropriately combined. Note that identical or corresponding parts in the drawings will be designated by the same reference numerals, and their description will not be repeated.

[0013] Embodiment 1. A control system 1 according to embodiment 1 will be described with reference to FIGS. 1 to 12. FIG. 1 is a diagram illustrating an example of the configuration of the control system 1 according to embodiment 1. As illustrated in FIG. 1, the control system 1 includes a plurality of electric devices 10A, 10B, and 10C and a server device 20. The plurality of electric devices 10A, 10B, and 10C each include a plurality of electric motors 12A, 12B, and 12C. Hereinafter, the electric devices 10A, 10B, and 10C will also be collectively referred to as electric devices 10. The electric motors 12A, 12B, and 12C will also be collectively referred to as electric motors 12. Note that if the electric device 10A is an example of a "first electric device" in the present disclosure, and the electric device 10B or the electric device 10C is an example of a "second electric device" in the present disclosure, the electric motor 12A is an example of a "first electric motor" in the present disclosure, and the electric motor 12B or the electric motor 12C is an example of a "second electric motor" in the present disclosure.

[0014] The electric device 10 (electric device 10A in the example of FIG. 1 ) includes a control device 11, an electric motor 12 (electric motor 12A in the example of FIG. 1 ), and a display 13. The control device 11 controls the electric motor 12 and the display 13. Note that the display 13 may be included in the configuration of the control device 11.

[0015] The electric motor 12 is a machine that converts electrical energy into mechanical energy. The electric motor 12 includes a sensor 125. The sensor 125 detects motor data related to the state of the electric motor 12. For example, the sensor 125 includes at least one of a current sensor that detects a current flowing through the electric motor 12, a voltage sensor that detects a voltage applied to the electric motor 12, a torque sensor that detects the torque of the electric motor 12, a sensor that detects the rotational position, rotational speed, or number of rotations of the electric motor 12 (e.g., an encoder 121 or a tachometer generator, which will be described later), a temperature sensor that detects the temperature of the electric motor 12, and a vibration sensor that detects vibrations of the electric motor 12. The sensor 125 may also detect environmental data related to the environment in which the electric motor 12 is installed. For example, the sensor 125 includes at least one of a temperature sensor that detects the temperature of the external environment in which the electric motor 12 is installed, a humidity sensor that detects the humidity of the external environment, an air pressure sensor that detects the air pressure of the external environment, a carbon dioxide concentration sensor that detects the carbon dioxide concentration of the external environment, and a pollutant concentration sensor that detects the pollutant concentration of the external environment.

[0016] Fig. 2 is a diagram showing an example of the configuration of the electric motor 12 included in the electric device 10 according to the first embodiment. The electric motor 12 is a machine that converts electrical energy into mechanical energy, and operates in accordance with control data from the control device 11. In the first embodiment, the electric motor 12 is a permanent magnet synchronous motor. As shown in Fig. 2, the electric motor 12 includes an encoder 121, a bearing 122, a stator 123, and a rotor 124.

[0017] The encoder 121 is an example of a sensor 125. The encoder 121 is a sensor for detecting the position or rotational speed of the rotor 124 during rotation and is used by the control device 11 to accurately control the electric motor 12. The bearings 122 support the rotor 124 and allow it to rotate smoothly. The bearings 122 reduce friction during rotation of the rotor 124, thereby achieving a long life for the rotor 124. The stator 123 is configured by windings wound around a stator core. The stator 123 generates a rotating magnetic field in the stator core by passing an AC current through the windings using power supplied from the control device 11 via the electric motor interface 105 (described later) based on the position information of the rotor 124 detected by the encoder 121. The rotor 124 is the rotating part of the electric motor 12 and is configured by attaching a permanent magnet and a shaft to the rotor core. The rotor 124 rotates in response to the rotating magnetic field generated in the stator 123.

[0018] In the electric motor 12 configured as described above, the rotor 124 rotates in synchronization with the rotating magnetic field generated by the stator 123 due to the interaction between the rotating magnetic field generated by the AC current flowing through the stator 123 and the magnetic field formed by the permanent magnet of the rotor 124. The rotor 124 rotates smoothly via the bearing 122, and the rotational position or rotational speed of the rotor 124 is detected by the encoder 121. The detected rotational position or rotational speed is input to the control device 11 via the electric motor interface 105. The control device 11 can accurately control the electric motor 12 by adjusting the power supplied to the stator 123 based on the rotational position or rotational speed obtained from the electric motor 12.

[0019] Returning to FIG. 1, the control device 11 includes an arithmetic unit 101 , a memory 102 , a storage device 103 , a communication interface 104 , a motor interface 105 , a display interface 106 , and a drive circuit 107 .

[0020] The arithmetic device 101 is a computing entity (computer) that controls actuators such as the electric motor 12 and the display 13 by executing various programs. The arithmetic device 101 may be configured, for example, as a microcontroller, a central processing unit (CPU), a micro processing unit (MPU), a tensor processing unit (TPU), or a graphics processing unit (GPU). The arithmetic device 101 has the function of performing various processes by executing programs, but some or all of these functions may be implemented using dedicated hardware circuits such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). The arithmetic device 101 is not limited to processors in the strict sense that perform processes using stored programs, such as a CPU, MPU, TPU, or GPU, but may also include hardwired circuits such as an ASIC or FPGA. Furthermore, the arithmetic device 101 is not limited to von Neumann computers such as a CPU or GPU, but may also be configured as non-von Neumann computers such as a quantum computer or an optical computer. The arithmetic device 101 can also be interpreted as a processing circuit that executes a predetermined process. The arithmetic device 101 may be configured as a single chip or multiple chips. Furthermore, the arithmetic device 101 and related processing circuits may be configured as multiple computers interconnected by wire or wirelessly via a local area network or a wireless network. The arithmetic device 101 and related processing circuits may be configured as a cloud computer that performs remote calculations based on input data and outputs the calculation results to another device located at a distance.

[0021] The memory 102 includes a storage area (for example, a working area) that stores program code or work memory when the arithmetic device 101 executes various programs. Examples of the memory 102 include volatile memories such as DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory), and non-volatile memories such as ROM (Read Only Memory) and flash memory.

[0022] The storage device 103 stores various programs and various data executed by the arithmetic device 101. The storage device 103 may be one or more non-transitory computer-readable media, or may be one or more computer-readable storage media. Examples of the storage device 103 include a hard disk drive (HDD) and a solid state drive (SSD).

[0023] The communication interface 104 is an interface configured to be able to communicate with the server device 20 and each of the other electric devices 10 (electric devices 10B and 10C in the example of FIG. 1 ) via wired communication or wireless communication. The arithmetic device 101 can transmit and receive data to and from the server device 20 and each of the other electric devices 10 via the communication interface 104.

[0024] The motor interface 105 is an interface configured to be able to communicate with the electric motor 12, which is the object of control. The arithmetic device 101 can send and receive data to and from the electric motor 12 via the motor interface 105.

[0025] Fig. 3 is a diagram showing an example of the configuration of the motor interface 105 included in the control device 11 according to Embodiment 1. As shown in Fig. 3, the motor interface 105 includes a data logger 151, a data buffer 152, and a data preprocessing unit 153.

[0026] The data logger 151 receives at least one of motor data relating to the state of the motor 12 and environmental data relating to the environment in which the motor 12 is installed, input from a sensor 125 provided on the motor 12. The data logger 151 stores the input data (motor data, environmental data) input from the sensor 125 in chronological order along with a timestamp. The input data from the sensor 125 is analog data. Therefore, the input data acquired by the data logger 151 is converted to digital data by an analog-to-digital converter (ADC). The data buffer 152 is a memory for temporarily storing the input data collected by the data logger 151. The data preprocessing unit 153 optimizes the input data so that it is suitable for analysis by the arithmetic device 101 by removing noise from the input data temporarily stored in the data buffer 152 and correcting or smoothing the input data. The input data from the sensor 125 optimized by the data preprocessing unit 153 is acquired by the arithmetic device 101.

[0027] In the motor interface 105 configured as described above, input data (motor data, environmental data) from the sensor 125 is acquired by the data logger 151, converted into digital data, and temporarily stored in the data buffer 152. The data buffer 152 serves to prevent input data overflow when input data from the sensor 125 is continuously collected. The input data collected by the data buffer 152 is subjected to noise removal or correction by the data preprocessing unit 153, and is output to the arithmetic device 101.

[0028] 1 , the display interface 106 is an interface configured to be able to communicate with the display 13, which is the object to be controlled. The arithmetic device 101 can send and receive data to and from the display 13 via the display interface 106.

[0029] The drive circuit 107 is a circuit that generates control data for controlling the electric motor 12. Fig. 4 is a diagram showing an example of the configuration of the drive circuit 107 provided in the control device 11 according to embodiment 1. As shown in Fig. 4, the drive circuit 107 includes a power supply 171, a converter 172, and an inverter 173.

[0030] The power supply 171 supplies the power necessary to operate the electric motor. The converter 172 converts the AC power supplied from the power supply 171 into DC power. The inverter 173 converts the DC power supplied from the converter 172 into AC power of a voltage and frequency desired by the calculation device 101, thereby generating power for driving the electric motor 12. The inverter 173 is operated by PWM (Pulse Width Modulation) control, and is capable of accurately controlling the voltage and frequency.

[0031] In the drive circuit 107 configured as described above, the operations of the converter 172 and the inverter 173 are adjusted according to the control of the arithmetic device 101. The inverter 173 is operated by PWM control, and outputs AC power of a voltage and frequency desired by the arithmetic device 101 as control data to the electric motor 12 via the electric motor interface 105. This enables the electric motor 12 to operate according to the control data from the control device 11.

[0032] In the electrical device 10 configured as described above, at least one of motor data relating to the state of the motor 12 and environmental data relating to the environment in which the motor 12 is installed is detected by the sensor 125. At least one of the motor data and environmental data detected by the sensor 125 is input as input data to the control device 11. Based on the input data, the control device 11 generates control data for appropriately controlling the motor 12 using the drive circuit 107, and outputs the control data to the motor 12, thereby operating the motor 12.

[0033] 1 , the server device 20 is communicably connected to the control devices 11 mounted on each of the plurality of electrical devices 10A, 10B, and 10C via a network. The server device 20 may be a server device disposed at the installation location of the electrical device 10, or may be a cloud-based server device disposed at a location different from the installation location of the electrical device 10. The server device 20 includes a calculation device 201, a memory 202, a storage device 203, and a communication interface 204.

[0034] The arithmetic device 201 has the same configuration as the arithmetic device 101 of the control device 11, and is a computing entity (computer) that controls each actuator by executing various programs. Like the arithmetic device 101, the arithmetic device 201 is configured with a microcontroller, a CPU, an MPU, a TPU, a GPU, an ASIC, an FPGA, or the like. The arithmetic device 201 can also be interpreted as a processing circuitry that executes predetermined processing.

[0035] The memory 202 has a configuration similar to that of the memory 102 of the control device 11, and includes a storage area (for example, a working area) for storing program code or work memory when the arithmetic device 201 executes various programs. The memory 202 is configured as a volatile memory or a non-volatile memory.

[0036] The storage device 203 has a configuration similar to that of the storage device 103 of the control device 11, and stores various programs executed by the arithmetic device 201, various data, etc. The storage device 203 may be one or more non-transitory computer readable media, or may be one or more computer readable storage media.

[0037] The communication interface 204 is an interface configured to be able to communicate with each of the plurality of electrical devices 10 via wired communication or wireless communication. The arithmetic device 201 is able to transmit and receive data to and from each of the plurality of electrical devices 10 via the communication interface 204.

[0038] In the control system 1 configured as described above, the control device 11 of each electrical device 10 is configured to estimate (predict) lifespan data related to the lifespan of the electric motor 12. Specifically, the storage device 103 of the control device 11 stores an estimation model 131 and a control program 132. The control program 132 includes a training program 132A and an estimation program 132B.

[0039] The estimation model 131 is trained (for example, by machine learning) to estimate lifetime data based on input data correlated with lifetime data related to the lifetime of the electric motor 12. Before shipping the electric device 10, the manufacturer of the electric device 10 may prepare input data and lifetime data of an electric motor 12 having the same performance as the electric motor 12 (for example, an electric motor in the same lot) as the electric motor 12 as training data for training the electric motor 12 to be installed in the electric device 10, and may use the training data to train the estimation model 131 to be installed in the electric device 10 to be shipped.

[0040] The input data includes at least one of motor data related to the state of the motor 12, environmental data related to the environment in which the motor 12 is installed, abnormality data related to an abnormality in the motor 12, and feedback data related to feedback on the operation of the motor. The lifespan data includes the timing (e.g., date and time) when the lifespan of the motor 12 will end, or the period from when the lifespan of the motor 12 was estimated to when the lifespan of the motor 12 will end.

[0041] The motor data includes data related to the state of the motor 12 detected by the sensor 125. For example, the motor data includes at least one of the current flowing through the motor 12, the voltage applied to the motor 12, the torque of the motor 12, the rotational position of the motor 12, the rotational speed of the motor 12, the number of revolutions of the motor 12, the temperature of the motor 12, and the vibration of the motor 12. The control device 11 can acquire the motor data from the sensor 125 via the motor interface 105. Note that the motor data is not limited to the data detected by the sensor 125 and may include, for example, identification information of the motor 12 (such as the product number or serial number of the motor 12). In this case, the control device 11 may store the identification information of the motor 12 in the storage device 103 in advance. The motor data may include any data indicating the state of the motor 12.

[0042] The environmental data includes data about the environment in which the electric motor 12 is installed, detected by the sensor 125. For example, the environmental data includes at least one of the temperature, humidity, atmospheric pressure, carbon dioxide concentration, and pollutant concentration of the external environment in which the electric motor 12 is installed. The control device 11 can acquire the environmental data from the sensor 125 via the motor interface 105. The environmental data is not limited to the data detected by the sensor 125, but may also include at least one of a load state indicating the load on the electric motor 12 and location information indicating the location of the electric motor 12. In this case, the control device 11 may store the load state and location information of the electric motor 12 in the storage device 103 in advance. The environmental data may include any data indicating the state of the environment in which the electric motor 12 is installed. The environmental data may be detected not only by the sensor 125 but also by other Internet of Things (IoT) devices connected to the control device 11. In this case, the control device 11 may acquire the environmental data from the other IoT devices via the communication interface 104.

[0043] The abnormality data includes data indicating an abnormality that has occurred in the electric motor 12 or a failure of some actuator included in the electric motor 12. The abnormality data may include data indicating the specific details of the abnormality or failure, or may include identification information (for example, an abnormality code or a failure code) corresponding to the details of the abnormality or failure. The control device 11 can acquire the abnormality data from the electric motor 12 via the electric motor interface 105.

[0044] The feedback data includes feedback information from a user regarding the operation of the electric motor 12. The user may be a user of the electric motor 12, a manager of the electric motor 12, or a worker who maintains the electric motor 12. The feedback data may be input to the control device 11 by the user using an input device such as a keyboard or a mouse (not shown), or may be input from the server device 20 via the communication interface 104.

[0045] The training program 132A is a program describing a process (training process) for training the estimation model 131 so that the estimation model 131 estimates lifespan data based on input data. The arithmetic device 101 can train the estimation model 131 by executing the training program 132A.

[0046] The estimation program 132B is a program describing a process (estimation process) for estimating lifespan data based on input data using the estimation model 131. By executing the estimation program 132B, the arithmetic device 101 can estimate lifespan data based on the acquired input data.

[0047] In this way, the control device 11 of the electrical equipment 10 is configured to use the trained estimation model 131 to estimate the life data of the electric motor 12 based on at least one input data of the electric motor data, the environmental data, the abnormality data, and the feedback data.

[0048] The control device 11 controls the electric motor 12 based on the life data of the electric motor 12 estimated using the estimation model 131. Specifically, the control device 11 controls the electric motor 12 so that the electric motor 12 operates over a life corresponding to the life data of the electric motor 12 estimated using the estimation model 131.

[0049] For example, the control device 11 changes parameters such as the rotational speed, torque, voltage, and current of the electric motor 12 to change the operating mode of the electric motor 12 or adjust the operating conditions of the electric motor 12. When the control device 11 detects an increase in load, the control device 11 increases the torque of the electric motor 12 to accommodate the increase in load, thereby preventing a shortening of the lifespan of the electric motor 12. For example, when deterioration of the electric motor 12 progresses, the control device 11 may slow the progression of deterioration of the electric motor 12 by limiting the operation of the electric motor 12. The control device 11 controls the electric motor 12 so that the electric motor 12 operates for the lifespan estimated using the estimation model 131 by changing parameters such as the rotational speed, torque, voltage, and current of the electric motor 12 to reduce power consumption. When the control device 11 detects an abnormality in the electric motor 12, the control device 11 takes emergency measures, such as stopping the operation of the electric motor 12, to prevent a shortening of the lifespan of the electric motor 12.

[0050] Furthermore, the control device 11 executes processing based on the following algorithm as control for improving the energy efficiency of the electrical device 10 and preventing the lifespan from being shortened.

[0051] For example, the control device 11 acquires motor data and environmental data of the electric motor 12 to calculate the amount of power consumption. The control device 11 compares the amount of power consumption with the output performance of the electric motor 12 to identify conditions under which efficiency decreases. The control device 11 minimizes power consumption by adjusting or changing the operating mode under conditions under which efficiency decreases. This allows the control device 11 to optimize the power consumption during operation of the electric motor 12 in real time, which can lead to savings on electricity bills or reductions in carbon dioxide emissions. Furthermore, the control device 11 can maximize the performance of the electric motor 12 and avoid unnecessary energy consumption. By incorporating environmental data into the calculation of energy efficiency, the control device 11 can flexibly respond to environmental fluctuations.

[0052] For example, the control device 11 detects an abnormal pattern or a sign of an abnormality based on the motor data of the electric motor 12. The control device 11 determines a preventive measure corresponding to the abnormal pattern or the sign of an abnormality, such as changing the operation mode or stopping operation, and executes the determined preventive measure. This allows the control device 11 to execute the preventive measure before an abnormality occurs in the electric motor 12, thereby significantly reducing the breakdown or downtime of the electric motor 12. This allows the control device 11 to reduce the cost of repairing or replacing the electric motor 12 and improve the safety or reliability of the entire control system.

[0053] For example, the control device 11 estimates the degree of wear or deterioration of components constituting the electric motor 12 based on motor data or environmental data of the electric motor 12. The control device 11 predicts the timing of necessary maintenance or inspection based on the estimated degree of wear or deterioration of the components. The control device 11 executes processing to optimize a user's maintenance schedule based on the predicted timing of maintenance or inspection. As a result, the control device 11 accurately estimates the degree of wear or deterioration of the components of the electric motor 12, thereby avoiding excessive maintenance and enabling appropriate allocation of resources. Furthermore, the control device 11 optimizes the user's schedule based on the predicted timing of maintenance or inspection, thereby reducing long-term operating costs or labor. As a result, the control device 11 can extend the life of the electric motor 12 and maintain its performance sustainably.

[0054] By combining the above-described algorithms, the control device 11 can significantly contribute to efficient operation of the electric motor 12, high reliability of the electric motor 12, and improved cost efficiency of the electric motor 12. Specifically, the control device 11 can achieve effects in three aspects: improved energy efficiency, prevention of failures, and optimization of maintenance. By employing such algorithms, the control device 11 can realize optimization of the overall life cycle of the electric motor 12.

[0055] Furthermore, the control device 11 controls the electric motor 12 so that the electric motor 12 operates over a lifetime corresponding to the lifetime data of the electric motor 12 estimated using the estimation model 131. Then, the control device 11 periodically estimates the lifetime data of the electric motor 12 using the estimation model 131 based on input data, such as electric motor data and environmental data, obtained when the electric motor 12 is controlled. The control device 11 controls the electric motor 12 so that the electric motor 12 operates over a lifetime corresponding to the periodically estimated lifetime data. This allows the control device 11 to control the electric motor 12 so that the lifetime is periodically extended over time based on input data collected in real time during daily operation, thereby improving the efficiency and lifetime of the electric motor 12. Furthermore, the control device 11 can quickly respond to fluctuations in the operating state of the electric motor 12 or the external environment and constantly perform optimal control, thereby improving the durability of the electric motor 12 and reducing operating costs.

[0056] Furthermore, the control device 11 is configured to be able to retrain the estimation model 131 based on input data correlated with lifespan data on the lifespans of other electric motors 12 than the electric motor 12 controlled by the control device 11. Hereinafter, the electric motor 12 controlled by the control device 11 will also be referred to as the "controlled electric motor." The controlled electric motor is an example of the "first electric motor." The other electric motor 12 not controlled by the control device 11 will also be referred to as the "non-controlled electric motor." The non-controlled electric motor is an example of the "second electric motor."

[0057] For example, the control device 11 that controls the electric motor 12A (first electric motor) is configured to use a trained estimation model 131 to estimate the lifespan data of the electric motor 12A based on input data (first input data) such as motor data, environmental data, abnormality data, and feedback data that are correlated with the lifespan data (first lifespan data) related to the lifespan of the electric motor 12A, and to retrain the estimation model 131 based on input data (second input data) such as motor data, environmental data, abnormality data, and feedback data that are correlated with the lifespan data (second lifespan data) related to the lifespan of the electric motors 12B and 12C (second electric motors) that are not to be controlled.

[0058] This allows the control device 11 to improve the estimation accuracy of the estimation model 131 by taking into account not only the lifespan of the controlled electric motor 12 but also that of the non-controlled electric motor 12. For example, the control device 11 retrains the estimation model 131 using training data that is a set of motor data used in the non-controlled electric motor 12 and the actual lifespan when the non-controlled electric motor 12 is controlled using the motor data. The control device 11 retrains the estimation model 131 using training data that is a set of environmental data related to the environment in which the non-controlled electric motor 12 is installed and the actual lifespan of the non-controlled electric motor 12 operating in that environment. The control device 11 also retrains the estimation model 131 using training data that is a set of abnormality data indicating an abnormality that has occurred in the non-controlled electric motor 12 and the actual lifespan of the non-controlled electric motor 12 operating in a state in which the abnormality has occurred.

[0059] In this way, the control device 11 can improve the estimation accuracy of the estimation model 131 by retraining the estimation model 131 using experience with the non-controlled electric motor 12. The control device 11 can always use the latest estimation model 131 to appropriately control the electric motor 12 so as to extend the life of the electric motor 12 by retraining the estimation model 131 based on new motor data, environmental data, and abnormality data used in the non-controlled electric motor 12. For example, the control device 11 can accurately predict the life of the electric motor 12 using the estimation model 131 retrained based on experience with the non-controlled electric motor 12, based on environmental data indicating temperature changes or humidity changes, and can control the electric motor 12 so that the electric motor 12 operates for the estimated life. For example, if the electric motor 12 generates abnormal vibrations, the control device 11 can accurately predict the life of the electric motor 12 using the estimation model 131 retrained based on experience with the non-controlled electric motor 12, based on motor data indicating the vibrations. Furthermore, even if a new abnormality occurs, the control device 11 can accurately predict the lifespan using the retrained estimation model 131, and can control the electric motor 12 so that the electric motor 12 operates for the estimated lifespan. This allows the control device 11 to prevent downtime caused by a breakdown of the electric motor 12. In this way, the user can accurately estimate the lifespan of the electric motor 12 using the estimation model 131 without misjudging the lifespan of the electric motor 12, and can therefore appropriately maintain the electric motor 12 in consideration of the estimated lifespan.

[0060] Fig. 5 is a diagram showing the functional configuration of the control device 11 according to embodiment 1. As shown in Fig. 5, the control device 11 includes a calculation unit 110 corresponding to the functional configuration of the calculation device 101, a storage unit 130 corresponding to the functional configuration of the storage device 103, and an interface unit 150 corresponding to the functional configurations of the communication interface 104, the electric motor interface 105, and the display interface 106.

[0061] The storage unit 130 stores various data such as an estimation model 131 and a control program 132 (training program 132A, estimation program 132B).

[0062] The interface unit 150 acquires input data (motor data, environmental data, abnormality data, feedback data) from the electric motor 12 and outputs the acquired input data to the calculation unit 110. The interface unit 150 also acquires input data (motor data, environmental data, abnormality data, feedback data) and lifespan data from electric motors 12 that are not subject to control and outputs the acquired input data and lifespan data to the calculation unit 110. The interface unit 150 also outputs control data for controlling the electric motor 12, generated by the control unit 114, to the electric motor 12. The interface unit 150 also acquires image data for displaying an image on the display 13 from the calculation unit 110 and outputs the acquired image data to the display 13.

[0063] The calculation unit 110 includes a training unit 111 , an estimation unit 112 , a data acquisition unit 113 , a control unit 114 , and a display unit 115 .

[0064] The training unit 111 has a function of training the estimation model 131 using training data that is a set of input data and life data of a training electric motor 12 prepared in advance. Furthermore, the training unit 111 has a function of retraining the trained estimation model 131 based on life data related to the life of an electric motor 12 that is not subject to control and input data that is correlated with the life data.

[0065] The estimation unit 112 has a function of estimating lifespan data based on input data acquired by the data acquisition unit 113 using a trained or retrained estimation model 131 .

[0066] The data acquisition unit 113 has a function of acquiring and collecting input data acquired by the interface unit 150. The data acquisition unit 113 may adjust the frequency or range of input data collection depending on the operating state of the electric motor 12 or the external environment. This allows the control device 11 to efficiently collect only the necessary input data and optimally utilize the acquisition range of the data acquisition unit 113 or the interface unit 150.

[0067] The control unit 114 has a function of generating control data for controlling the electric motor 12 based on the life data estimated by the estimation unit 112 .

[0068] Display unit 115 has a function of generating image data for displaying an image on display 13. For example, display unit 115 generates image data for displaying information related to the lifespan data estimated by estimation unit 112 on display 13, and outputs the image data to display 13 via interface unit 150. Based on the image data from control device 11, display 13 displays an image indicating the timing (e.g., date and time) at which the lifespan of electric motor 12 will end, or the period from the timing at which the lifespan of electric motor 12 was estimated to the timing at which the lifespan of electric motor 12 will end.

[0069] In this way, the electric device 10 can accurately estimate the lifespan of the electric motor 12 and efficiently control the electric motor 12 so that it operates over the estimated lifespan by using the control device 11 having various functional units such as the training unit 111, the estimation unit 112, the data acquisition unit 113, the control unit 114, and the display unit 115. Furthermore, the electric device 10 can display information related to the lifespan data estimated by the control device 11 on the display 13, thereby clearly communicating the condition of the electric motor 12 to the user. This allows the user to appropriately review the timing of maintenance of the electric motor 12 and the operation policy.

[0070] As described above, the control device 11 performs supervised learning using training data that is a set of input data and lifespan data, which is ground truth data corresponding to the input data. Supervised learning is a technique that uses a data set of factors and results (labels) to learn the features of the training data and estimate the results from the input. Figure 6 is a diagram for explaining an overview of supervised learning.

[0071] As shown in FIG. 6, in the training phase, the control device 11 executes the training program 132A to train the estimation model 131 based on training data 180 including input 1 and input 2 (correct answer).

[0072] In the exploitation phase, the controller 11 uses the estimation model 131 to obtain an output based on the input 1 .

[0073] 7 is a diagram illustrating the input and output of supervised learning in the control device 11 according to the first embodiment. As shown in FIG. 7 , in the control device 11, input data correlated with the life data of the electric motor 12 is used as input 1. When training the estimation model 131 before the shipment of the electric device 10, the input data of the electric motor 12 used for training is used as input 1. When estimating the life using the estimation model 131 after the shipment of the electric device 10, the input data of the electric motor 12 to be controlled and mounted on the electric device 10 is used as input 1. When training the estimation model 131 after the shipment of the electric device 10, the input data of the electric motor 12 to be controlled is used as input 1.

[0074] The input data includes at least one of motor data relating to the state of the motor 12, environmental data relating to the environment in which the motor 12 is installed, abnormality data relating to abnormalities in the motor 12, and feedback data relating to feedback on the operation of the motor 12.

[0075] 7 and may include other data indicating the operating state or trend of the electric motor 12. For example, the input data may include at least one of the operating time of the electric motor 12, the amount of power consumption of the electric motor 12, the state of occurrence of an abnormality of the electric motor 12, the fluctuation trend of the power consumption of the electric motor 12, the fluctuation trend of the efficiency of the electric motor 12, and the frequency of occurrence of an abnormality of the electric motor 12.

[0076] The operating time of the electric motor 12 includes the continuous operating time or daily operating time of the electric motor 12. The power consumption of the electric motor 12 includes the power consumption of the electric motor 12 measured in real time or during a specific time period. The abnormality occurrence state of the electric motor 12 includes the presence or absence or frequency of various abnormalities detected through data analysis by the sensor or the control device 11. The trend of the operating time of the electric motor 12 includes an increase or decrease in the operating time of the electric motor 12 by month, season, and region. The fluctuation trend of the power consumption of the electric motor 12 includes an increase or decrease in the power consumption of the electric motor 12 during a specific period or a change in the power consumption of the electric motor 12 during peak hours. The fluctuation trend of the efficiency of the electric motor 12 includes the results of an analysis of the efficiency of the electric motor 12 over time, and particularly includes a decrease in the efficiency of the electric motor 12 due to aging or component wear. The frequency of abnormalities occurring in the electric motor 12 includes a change in the number or type of abnormalities occurring in the electric motor 12 during a specific period.

[0077] In the control device 11, the life data of the electric motor 12 corresponding to the input data of the input 1 is used as the input 2 which is the correct data.

[0078] In the control device 11, life data of the electric motor 12 to be controlled is used as an output.

[0079] 8 is a diagram showing an overview of the training process executed by the control device 11 (the calculation unit 110) according to embodiment 1. The control device 11 is capable of exchanging data with each of the training program storage unit 136 and the estimation model storage unit 137. The training program storage unit 136 and the estimation model storage unit 137 are realized by the storage unit 130 of the control device 11.

[0080] 8 , the control device 11 includes a data acquisition unit 113 and a training unit 111. The control device 11 executes a training program 132A stored in a training program storage unit 136, thereby training or retraining an estimation model 131 based on training data 180 including an input 1 and an input 2 (correct answer).

[0081] The data acquisition unit 113 acquires training data 180 including input 1 and input 2 (correct answer). The training unit 111 uses the training data 180 including input 1 and input 2 (correct answer) acquired by the data acquisition unit 113 to train the estimation model 131 so as to estimate life data of the electric motor 12 as an output from input 1. The training unit 111 stores the generated estimation model 131 in the estimation model storage unit 137.

[0082] 9 is a diagram showing the configuration of a neural network. The training unit 111 trains the estimation model 131 by supervised learning, for example, in accordance with the neural network model.

[0083] A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer, or two or more layers.

[0084] 9 shows a three-layer neural network. It shows a configuration with three inputs and three outputs. When multiple inputs are input to input layers X1, X2, and X3, the values ​​are multiplied by weights w11 to w16, and the resulting values ​​are input to intermediate layers Y1 and Y2. The resulting values ​​are then further multiplied by weights w21 to w26, and output from output layers Z1, Z2, and Z3. The output results vary depending on the values ​​of the weights w11 to w16 and w21 to w26.

[0085] The neural network performs supervised learning based on training data 180 including input 1 and input 2 (correct answer) acquired by the data acquisition unit 113. That is, the neural network learns by inputting input 1 to the input layer and adjusting the weights so that the result output from the output layer approaches input 2 (correct answer).

[0086] The training unit 111 trains the estimation model 131 by performing supervised learning as described above.

[0087] In this way, by incorporating machine learning to train the estimation model 131, the control device 11 can learn complex data relationships and improve the accuracy of estimating lifespan data.

[0088] 10 is a diagram showing an outline of the estimation process executed by the control device 11 according to embodiment 1. The control device 11 is capable of exchanging data with the estimation model storage unit 137.

[0089] 10 , the control device 11 includes a data acquisition unit 113 and an estimation unit 112. The data acquisition unit 113 acquires input data of the electric motor 12 to be controlled as an input 1. The estimation unit 112 uses an estimation model 131 read out from an estimation model storage unit 137 to output life data of the electric motor 12 to be controlled as an output based on the input data acquired by the data acquisition unit 113.

[0090] Fig. 11 is a flowchart relating to processing executed by the control device 11 according to the first embodiment in a training phase. The training phase is assumed to be, for example, a stage in which the estimation model 131 is trained before the shipment of the electrical device 10. Fig. 11 illustrates processing executed by the arithmetic device 101 of the control device 11. The arithmetic device 101 executes the control program 132 (training program 132A) to perform each of the processing steps illustrated in Fig. 11. Note that in Fig. 11, "S" is used as an abbreviation for "STEP."

[0091] 11 , the control device 11 determines whether training data including input 1 and input 2 (correct answer) has been input via the communication interface 104 or the motor interface 105 (S1). If no training data has been input (NO in S1), the control device 11 ends this processing. On the other hand, if training data has been input (YES in S1), the control device 11 acquires the training data via the data acquisition unit 113 (S2).

[0092] The control device 11 trains (learns, generates) the estimation model 131 by performing supervised learning using the training data 180 using the training unit 111 (S3). The control device 11 stores the generated estimation model 131 in the storage device 103 (S4), and then ends this process.

[0093] Fig. 12 is a flowchart of processing executed by the control device 11 according to the first embodiment in the utilization phase. The utilization phase is assumed to be, for example, a stage in which the estimation model 131 is retrained after the shipment of the electrical device 10, and lifetime data is estimated using the estimation model 131. Fig. 12 shows processing executed by the arithmetic device 101 of the control device 11. The arithmetic device 101 executes the control program 132 (training program 132A, estimation program 132B) to perform each processing shown in Fig. 12. Note that in Fig. 12, "S" is used as an abbreviation for "STEP."

[0094] 12, the control device 11 determines whether or not training data 180 including input 1 and input 2 (correct answer) has been input via the communication interface 104 or the motor interface 105 (S11). In this case, the training data 180 includes, for example, lifespan data related to the lifespan of the non-controlled electric motor 12 and input data correlated with the lifespan data.

[0095] If the training data 180 is not input (NO in S11), the control device 11 proceeds to the process of S15. On the other hand, if the training data 180 is input (YES in S11), the control device 11 acquires the training data 180 by the data acquisition unit 113 (S12).

[0096] The control device 11 retrains (learns) the estimation model 131 by performing supervised learning based on the training data 180 using the training unit 111 (S13). The control device 11 may retrain the estimation model 131 in real time every time training data 180 is acquired, or may retrain the estimation model 131 every time a predetermined period of time has passed, such as when a predetermined amount of training data 180 is acquired. The control device 11 updates the estimation model 131 stored in the estimation model storage unit 137 to the retrained estimation model 131 (S14), and proceeds to the process of S15.

[0097] The control device 11 determines whether input data (input 1) has been input (S15). In this case, the input data includes motor data of the electric motor 12 to be controlled, environmental data, etc. If no input data has been input (NO in S15), the control device 11 proceeds to the process of S21. On the other hand, if input data has been input (YES in S15), the control device 11 acquires the input data via the data acquisition unit 113 (S16).

[0098] The control device 11 inputs the acquired input data to the estimation model 131 (S17). The control device 11 uses the estimation model 131 to estimate life data of the electric motor 12 to be controlled based on the input data (S18).

[0099] Based on the estimated lifespan data of the electric motor 12, the control device 11 generates control data for operating the electric motor over the lifespan corresponding to the lifespan data (S19). The control device 11 outputs the control data to the electric motor 12 via the electric motor interface 105 (S20).

[0100] The control device 11 determines whether a termination condition is met (S21). For example, the termination condition includes the elapse of a predetermined operating time of the electrical device 10 or the occurrence of a specific abnormality. If the termination condition is not met (NO in S21), the control device 11 returns to the process of S11. On the other hand, if the termination condition is met (YES in S21), the control device 11 ends this process.

[0101] As described above, in the electrical device 10, the estimation model 131, which has been trained to estimate lifespan data based on input data correlated with lifespan data related to the lifespan of the controlled electric motor 12, is retrained based on input data correlated with lifespan data related to the lifespan of the non-controlled electric motor 12, so that the estimation accuracy of the estimation model 131 can be improved by taking into account the lifespan of not only the controlled electric motor 12 but also the non-controlled electric motor 12. This allows the user to accurately estimate the lifespan of the electric motor 12 using the estimation model 131, and therefore allows the user to appropriately maintain the electric motor 12 by taking into account the estimated lifespan.

[0102] The control device 11 periodically and continuously retrains and updates the estimation model 131, thereby enabling the electric motor 12 to always operate based on the latest technology and input data. This allows the control device 11 to flexibly adapt the electric motor 12 to changes in the external environment in which the electric motor 12 is installed or changes in the operating state of the electric motor 12. The control device 11 can quickly adapt the electric motor 12 to unknown changes in the operating environment or the occurrence of unpredictable abnormalities, thereby reducing the risk of failure of the electric motor 12 and extending the lifespan of the electric motor 12. The control device 11 can prevent events that shorten the lifespan of the electric motor 12, such as overheating or overload, and achieve stable operation. The control device 11 precisely controls the operation of the electric motor 12 by applying advanced control technology using the estimation model 131, thereby reducing energy consumption and extending the lifespan of the electric motor 12.

[0103] Second Embodiment A control system 1A according to a second embodiment will be described with reference to Figures 13 and 14. Note that, in the following, only the parts of the control system 1A according to the second embodiment that are different from the control system 1A according to the first embodiment will be described.

[0104] Fig. 13 is a diagram showing the configuration of a control system 1A according to embodiment 2. As shown in Fig. 13, in the control system 1A according to embodiment 2, a server device 20A, which is an example of a "control device," is configured to estimate (predict) lifespan data related to the lifespan of the electric motors 12 mounted in each of the plurality of electrical devices 10.

[0105] Specifically, the storage device 203 of the server device 20A stores an estimation model 231 and a control program 232. The control program 232 includes a training program 232A and an estimation program 232B. The estimation model 231 has the same configuration and function as the estimation model 131 stored in the storage device 103 of the control device 11 according to the first embodiment. The training program 232A has the same configuration and function as the training program 132A stored in the storage device 103 of the control device 11 according to the first embodiment. The estimation program 232B has the same configuration and function as the estimation program 132B stored in the storage device 103 of the control device 11 according to the first embodiment.

[0106] 14 is a diagram showing the functional configuration of server device 20A (control device) according to embodiment 2. As shown in Fig. 14, server device 20A includes a calculation unit 210 corresponding to the functional configuration of calculation unit 201, a storage unit 230 corresponding to the functional configuration of storage device 203, and an interface unit 250 corresponding to the functional configuration of communication interface 204.

[0107] The storage unit 230 stores various data such as an estimation model 231 and a control program 232 (training program 232A, estimation program 232B).

[0108] The interface unit 250 acquires input data (motor data, environmental data, abnormality data, feedback data) from the electric motors 12 mounted in each of the plurality of electric devices 10, and outputs the acquired input data to the calculation unit 210. The interface unit 250 also outputs life data of the electric motors 12 estimated by the estimation unit 212 to the electric devices 10 in which the electric motors 12 are mounted.

[0109] The calculation unit 210 includes a training unit 211 , an estimation unit 212 , and a data acquisition unit 213 .

[0110] The training unit 211 has a function of training the estimation model 231 using training data that is a set of input data and lifespan data of the electric motor 12 for training prepared in advance. Furthermore, the training unit 211 has a function of retraining the trained estimation model 131 based on lifespan data related to the lifespan of the electric motor 12 and input data correlated with the lifespan data.

[0111] The estimation unit 212 has a function of estimating lifespan data based on input data acquired by the data acquisition unit 213 using the trained or retrained estimation model 131 .

[0112] The data acquisition unit 213 has a function of acquiring and collecting input data acquired by the interface unit 250 .

[0113] In the control system 1A configured as described above, the server device 20A stores an estimation model 231 for predicting the lifetime of each of the electric motors 12 of the plurality of communicatively connected electric devices 10. When the server device 20A receives input data from one of the plurality of electric devices 10, the server device 20A uses the estimation model 231 to estimate lifetime data of the electric motor 12 installed in the electric device 10 based on the received input data, and outputs the estimated lifetime data to the electric device 10.

[0114] For example, server device 20A estimates lifetime data of electric motor 12A (first electric motor) included in electric device 10A (first electric device) based on input data (first input data) such as motor data, environmental data, abnormality data, and feedback data that are correlated with lifetime data (first lifetime data) related to the lifetime of electric motor 12A, and outputs the estimated lifetime data to electric device 10A. In this way, electric device 10A that has acquired the lifetime data estimated by server device 20A can control electric motor 12A based on the lifetime data so that electric motor 12A operates over the estimated lifetime.

[0115] The server device 20A may store a plurality of estimation models 231 used by each of a plurality of electric devices 10 connected to the server device 20A so as to be able to communicate with each other. In this case, the server device 20A may estimate the life data of the electric motor 12 mounted on the electric device 10 using the estimation model 231 corresponding to the electric device 10 from which the server device 20A has acquired input data. Alternatively, the server device 20A may store one estimation model 231 that is used in common by the plurality of electric devices 10 connected to the server device 20A so as to be able to communicate with each other. In this case, the server device 20A may estimate the life data of the electric motor 12 mounted on the electric device 10 using the one estimation model 231 from which the server device 20A has acquired input data.

[0116] Furthermore, the server device 20A can retrain the estimation model 231 based on input data acquired from each of the plurality of communicatively connected electric devices 10. For example, if the server device 20A stores a plurality of estimation models 231 used by the plurality of electric devices 10, when retraining one estimation model 231, the server device 20A retrains the one estimation model 231 based on input data acquired from an electric device 10 other than the electric device 10 that uses the one estimation model 231.

[0117] For example, server device 20A retrains estimation model 231, which is used to estimate lifetime data (first lifetime data) of electric motor 12A (first electric motor) included in electric device 10A (first electric device), based on input data (second input data) acquired from other electric devices 10B and 10C (second electric devices). In this way, server device 20A can improve the estimation accuracy of estimation model 231 by retraining estimation model 231 used in electric device 10A using the experiences of other electric devices 10B and 10C.

[0118] Alternatively, when the server device 20A stores one estimation model 231 that is commonly used by a plurality of electric devices 10, when retraining the one estimation model 231, the server device 20A retrains the one estimation model 231 based on input data acquired from each of the plurality of electric devices 10. In this way, the server device 20A can improve the estimation accuracy of the estimation model 231 by retraining the estimation model 231 that is commonly used by the electric devices 10 using the experience of all of the electric devices 10.

[0119] In this way, in the control system 1A, data is transmitted and received between the server device 20A on the cloud and the multiple electric devices 10, and the server device 20A can efficiently operate and manage the multiple electric devices 10 by periodically and continuously retraining and updating the estimation model 231 based on input data acquired from each electric device 10. Furthermore, the server device 20A retrains the estimation model 231 by machine learning that takes into account anomalies that occur in other electric devices 10, and therefore, a feedback loop can be established between each of the multiple electric devices 10 and the server device 20A (estimation model 231).

[0120] The server device 20A can accumulate and share experience and knowledge not only of individual electric devices 10 but also of the control system 1A as a whole. That is, the server device 20A can utilize experience or lessons learned from one electric device 10 to estimate the life span and optimize the control of other electric devices 10.

[0121] Furthermore, when a new electric device 10 is connected to the control system 1A, the server device 20A can estimate the lifespan of the electric motor 12 installed in the new electric device 10 using the estimation model 231, which has high estimation accuracy from the initial connection stage, based on the results of the federated learning performed up to that point, and can efficiently control the new electric motor 12. As a result, the new electric device 10 can be quickly incorporated into the control system 1A and is expected to operate stably. In this way, by utilizing federated learning on the cloud, it is possible to achieve not only improved performance of each electric device 10 but also improved efficiency and reliability of the entire system.

[0122] Furthermore, the server device 20A comprehensively manages and retrains the estimation model 231 commonly used by each of the multiple electrical devices 10, thereby enabling the server device 20A to quickly apply the estimation model 231 to abnormalities that occur in the multiple electrical devices 10 and continuously optimize the performance of the entire system.

[0123] Note that, because a large amount of input data is transmitted to the server device 20A in real time from each of the multiple electrical devices 10, the server device 20A analyzes the large amount of input data in real time to retrain the estimation model 231. For this reason, the server device 20A may analyze the input data in real time at high speed using a quantum computer.

[0124] The electrical device 10 in the first and second embodiments described above can be applied to an air conditioner. In this case, the sensor 125 detects data such as the indoor temperature, humidity, the operating mode of the air conditioner, the set temperature, and outdoor weather information. The electric motor 12 includes at least one of a compressor, an outdoor unit fan, and an indoor unit fan mounted on the air conditioner. These are key devices for maintaining efficient operation of the air conditioner.

[0125] The control device 11 (or the server device 20A) executes processing based on the following algorithm as control for improving the energy efficiency of the electrical device 10, which is an air conditioner, and preventing the lifespan from being shortened.

[0126] For example, the control device 11 acquires motor data (room temperature, set temperature, outside temperature, etc.) of the motor 12 and calculates the amount of power consumption. The control device 11 compares the amount of power consumption with the low heating performance of the air conditioner to identify conditions under which efficiency decreases. The control device 11 minimizes power consumption by adjusting or changing the operating mode (cooling, dehumidification, fan mode, etc.) under conditions under which efficiency decreases. This allows the control device 11 to optimize the power consumption during operation of the air conditioner in real time, which can lead to savings on electricity bills or reductions in carbon dioxide emissions. Furthermore, the control device 11 can automatically select the most efficient operating mode based on the set temperature or outside temperature.

[0127] For example, the control device 11 detects an abnormal pattern or a sign of an abnormality, such as a clogged filter or a decrease in refrigerant, based on the motor data of the motor 12. The control device 11 determines a preventive measure corresponding to the abnormal pattern or the sign of an abnormality, such as changing the operating mode or stopping operation, and executes the determined preventive measure. In this way, the control device 11 can detect abnormalities, such as a clogged filter or a decrease in refrigerant, early and provide preventive measures for these abnormalities, thereby extending the lifespan of the air conditioner and maintaining its performance.

[0128] For example, the control device 11 estimates the degree of filter contamination and the degree of wear or aging of components constituting the motor 12 based on the motor data of the motor 12. The control device 11 predicts the timing of necessary maintenance or inspection based on the estimated degree of wear or aging of the components. The control device 11 executes processing to optimize the user's maintenance schedule based on the predicted timing of maintenance or inspection. In this way, by accurately estimating the time to replace the filter and the degree of wear or aging of components of the motor 12, the control device 11 can avoid excessive maintenance and appropriately allocate resources. Furthermore, by optimizing the user's schedule based on the predicted timing of maintenance or inspection, the control device 11 can reduce long-term operating costs or labor.

[0129] The electric device 10 in the first and second embodiments described above may be applied to a vehicle. In this case, the sensor 125 may detect motor data and environmental data (temperature, humidity, road surface conditions, etc.) using sensors installed in various parts of the vehicle. The electric motor 12 includes at least one of an electric motor and other electric devices (power steering, brakes, air conditioning system, etc.) installed in the vehicle.

[0130] The control device 11 (or the server device 20A) executes processing based on the following algorithm as control for improving the energy efficiency of the electric device 10, which is a vehicle, and preventing the lifespan from being shortened.

[0131] For example, the control device 11 acquires electric motor data from each part of the vehicle in real time while the vehicle is traveling. The control device 11 controls the vehicle's electric motor 12 based on the electric motor data. As a result, the control device 11 can reduce fuel consumption or power consumption by efficiently operating the electric motor 12 installed in the vehicle. The control device 11 can detect signs of abnormalities or malfunctions in the vehicle early and prevent major malfunctions through proactive response. The control device 11 can optimize the timing of maintenance based on an estimate of wear or aging of vehicle components, thereby achieving long-term cost reductions or a longer vehicle lifespan.

[0132] Furthermore, the electrical device 10 in the above-described first and second embodiments may also be applied to electrical devices or related devices other than air conditioners and vehicles. For example, when the electrical device 10 is applied to an aircraft or a drone, the control device 11 can improve energy efficiency in the electrical system, detect abnormalities, or optimize maintenance and management, and monitor the performance or wear level of the electric motor 12 based on flight data, thereby improving safety or reducing costs.

[0133] Furthermore, when the electrical equipment 10 is applied to industrial machinery, the control device 11 can improve the efficiency of the electric motor 12 of a large industrial robot or a manufacturing line, detect abnormalities, or optimize maintenance and management, and by analyzing operation data of the production line, can detect overload or wear of the machine early, reducing downtime and realizing efficient production activities.

[0134] When the electrical equipment 10 is applied to a home appliance other than an air conditioner, such as a water heater, refrigerator, washing machine, ventilation fan, air purifier, or vacuum cleaner, the control device 11 can improve the energy efficiency of the home appliance, prevent breakdowns, or optimize maintenance, and by monitoring usage patterns or degree of wear, can suggest appropriate maintenance timing or energy-saving modes to consumers.

[0135] Furthermore, the control system 1 is not limited to connecting the same type of electrical appliances 10, but may connect multiple types of electrical appliances 10, such as home appliances, vehicles, aircraft, or industrial machinery. Furthermore, energy-related devices such as storage batteries and generators (solar, wind, etc.) are also included in the control system 1, so that input data from a wide variety of electric motors 12 is aggregated. This allows the control device 11 to grasp the operating status or trends of the entire system, and enables coordinated control even between different types of electrical appliances 10.

[0136] As described above, by applying the algorithms of the control device 11 to a wide variety of electrical devices 10, efficient operation, high reliability, or improved cost-effectiveness can be achieved depending on the application or environment of each electrical device 10.

[0137] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present disclosure is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0138] 1, 1A Control system, 10, 10A, 10B, 10C Electrical equipment, 11 Control device, 12, 12A, 12B, 12C Electric motor, 13 Display, 20, 20A Server device, 101, 201 Arithmetic device, 102, 202 Memory, 103, 203 Storage device, 104, 204 Communication interface, 105 Electric motor interface, 106 Display interface, 107 Drive circuit, 110, 210 Arithmetic unit, 111, 211 Training unit, 112, 212 Estimation unit, 113, 213 Data acquisition unit, 114 Control unit, 115 Display unit, 121 Encoder, 122 Bearing, 123 Stator, 124 Rotor, 125 Sensor, 130, 230 Memory unit, 131, 231 Estimation model, 132, 232 Control program, 132A, 232A training program, 132B, 232B estimation program, 136 training program storage unit, 137 estimation model storage unit, 150, 250 interface unit, 151 data logger, 152 data buffer, 153 preprocessing unit, 171 power supply, 172 converter, 173 inverter, 180 training data.

Claims

1. A control device for controlling an electric motor, comprising: a data acquisition unit that acquires first input data correlated with first life data regarding the life of a first electric motor; an estimation unit that estimates the first life data based on the first input data acquired by the data acquisition unit, using an estimation model trained to estimate the first life data based on the first input data; and a training unit that retrains the estimation model based on second input data correlated with second life data regarding the life of a second electric motor.

2. The control device according to claim 1, wherein the first input data includes at least one of first electric motor data regarding the state of the first electric motor, first environment data regarding the environment in which the first electric motor is installed, first abnormality data regarding an abnormality of the first electric motor, and first feedback data regarding feedback on the operation of the first electric motor.

3. The control device according to claim 1 or 2, wherein the second input data includes at least one of second electric motor data regarding the state of the second electric motor, second environment data regarding the environment in which the second electric motor is installed, second abnormality data regarding an abnormality of the second electric motor, and second feedback data regarding feedback on the operation of the second electric motor.

4. The control device according to any one of claims 1 to 3, wherein the estimation unit estimates the first life data based on the first input data acquired by the data acquisition unit, using the retrained estimation model.

5. The control device according to any one of claims 1 to 4, further comprising a control unit that controls the first electric motor based on the first life data estimated by the estimation unit.

6. The control device according to claim 5, wherein the control unit controls the first electric motor so that the first electric motor operates over a life corresponding to the first life data.

7. The control device according to any one of claims 1 to 6, further comprising a display unit that causes information regarding the first life data estimated by the estimation unit to be displayed on a display.

8. An electrical device comprising the control device according to any one of claims 1 to 7 and the first electric motor.

9. A control method for controlling an electric motor by a computer, the method comprising: obtaining first input data correlated with first life data regarding the life of a first electric motor as a process executed by the computer; estimating the first life data based on the first input data obtained in the obtaining step using an estimation model trained to estimate the first life data based on the first input data; and retraining the estimation model based on second input data correlated with second life data regarding the life of a second electric motor.

10. A control program for controlling an electric motor, the program causing a computer to execute: obtaining first input data correlated with first life data regarding the life of a first electric motor; estimating the first life data based on the first input data obtained in the obtaining step using an estimation model trained to estimate the first life data based on the first input data; and retraining the estimation model based on second input data correlated with second life data regarding the life of a second electric motor.

11. A control system for controlling an electric motor, the system comprising: a plurality of electrical devices including a first electrical device and a second electrical device; and a server device communicably connected to the plurality of electrical devices, the server device including: a data acquisition unit that acquires first input data correlated with first life data regarding the life of a first electric motor included in the first electrical device; an estimation unit that estimates the first life data based on the first input data acquired by the data acquisition unit using an estimation model trained to estimate the first life data based on the first input data; and a training unit that retrains the estimation model based on second input data correlated with second life data regarding the life of a second electric motor included in the second electrical device.

Images