Power supply module

Abstract

A power supply module (30) comprises: a power supply circuit (51) that generates output power by using a power generation voltage; and a control apparatus (100) that controls the power supply circuit (51) on the basis of a control parameter. The control apparatus (100): uses a digital twin of the power supply circuit (51) constructed using an estimation model (132) to estimate output power to be outputted from a virtual power supply circuit, on the basis of an input parameter of a power generation element (10) and a control parameter inputted to the power supply module; selects, on the basis of an estimated value of the output power, a control parameter that maximizes the output power; and controls the power supply circuit (51) on the basis of the selected control parameter.

Description

Power Supply

[0001] The present disclosure relates to a power supply module that outputs output power based on a generated voltage of a power generating element.

[0002] In recent years, energy harvesting elements that convert environmental energy (renewable energy) such as light energy, thermal energy, vibration energy, or electromagnetic wave energy into electric power have become known. For example, JP-A 2011-517277 (Patent Document 1) discloses an electromechanical generator that converts mechanical vibration energy into electric energy.

[0003] Special table 2011-517277 publication

[0004] The power generation element disclosed in Patent Document 1 can generate power using environmental energy, but a power supply circuit is required to efficiently use the generated voltage from the power generation element to output power to the charging element. Designing power supply circuits individually for various types of power generation elements increases design costs. In particular, while the market for energy harvesting elements is small, there are a wide variety of power generation methods. For this reason, there are significant barriers to designing power supply circuits suitable for such small-lot, high-mix energy harvesting elements, making it difficult for power supply circuits for energy harvesting elements to become widespread.

[0005] Furthermore, if a power supply circuit is individually designed to suit a specific power generation element, it is possible to configure a power supply circuit that can efficiently obtain output power corresponding to that specific power generation element, but the circuit configuration of the power supply circuit becomes complex and scalability such as adding other power generation elements is lost.On the other hand, if the power supply circuit is made versatile so that it can be adapted to various types of power generation elements, it becomes difficult to efficiently obtain output power according to each individual power generation element.

[0006] Therefore, an object of the present disclosure is to provide a power supply module that can efficiently obtain output power in accordance with various types of power generation elements.

[0007] A power supply module according to one embodiment of the present disclosure outputs output power based on a generated voltage of a power generation element. The power supply module includes a power supply circuit that generates output power using the generated voltage, and a control device that controls the power supply circuit based on control parameters. The control device uses a digital twin of the power supply circuit configured by an estimation model to estimate output power output from a virtual power supply circuit based on input parameters and control parameters of the power generation element input to the power supply module, selects control parameters that maximize the output power based on the estimated value of the output power, and controls the power supply circuit based on the selected control parameters.

[0008] According to the present disclosure, the control device uses a digital twin of the power supply circuit configured by an estimation model to estimate the output power output from the virtual power supply circuit based on the input parameters and control parameters of the power generation elements input to the power supply module, thereby enabling the selection of optimal control parameters for maximizing the output power output from the power supply circuit. This makes it possible to use the digital twin of the power supply circuit to select optimal control parameters for a power supply circuit having the required performance for various types of power generation elements, without having to design power supply circuits individually for the power generation elements, thereby providing a power supply module that can efficiently obtain output power for various types of power generation elements.

[0009] It is a block diagram of a power supply module according to an embodiment. It is a circuit diagram of a power supply module according to an embodiment. It is a diagram for explaining an example of an estimation model. It is a diagram for explaining another example of an estimation model. It is a diagram showing a simulation result related to training of an estimation model. It is a flowchart showing processing executed by a control device.

[0010] A power supply module 30 according to an embodiment will be described below with reference to the drawings. In the drawings, the same or corresponding components are denoted by the same reference numerals.

[0011] The main configuration of a power supply module 30 according to an embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram of the power supply module 30 according to an embodiment. As shown in Fig. 1, the power supply module 30, together with a power generation element 10 and a charging element 20, constitutes a power generation system 1. The power generation element 10 is connected to the input side of the power supply module 30. The charging element 20 is connected to the output side of the power supply module 30.

[0012] The power supply module 30 generates output power based on the voltage (generated voltage) generated by the power generating element 10 , and outputs the generated output power to the charging element 20 .

[0013] The power generating element 10 is an energy harvesting element that generates electricity using at least one of environmental energies: light energy, thermal energy, vibration energy, flow energy, rotational energy, electromagnetic wave energy, electric field energy, and magnetic field energy. For example, if the power generating element 10 is a photoelectric power generating element that generates electricity using light energy, the power generating element 10 converts light energy into electrical power using a phenomenon known as the photoelectric effect. If the power generating element 10 is a thermoelectric power generating element that generates electricity using thermal energy, the power generating element 10 extracts electrical power from thermoelectromotive force generated by the temperature difference between two types of metals or two types of thermoelectric conversion materials (e.g., p-type semiconductors and n-type semiconductors). If the power generating element 10 is an electrostatic vibration power generating element that generates electricity using vibration energy, the power generating element 10 converts vibration energy generated in a piezoelectric element into electrical power. If the power generating element 10 is an electromagnetic vibration power generating element that generates electricity using electromagnetic wave energy, the power generating element 10 extracts electrical power by vibrating a vibrator using environmental vibrations to change the magnetic flux. Note that the power generating element 10 is not limited to the above-described energy harvesting element, but may be other energy harvesting elements or a power generating element other than an energy harvesting element.

[0014] The charging element 20 is connected to a load circuit such as a sensor (not shown). The sensor is, for example, a wireless sensor that operates standalone in a factory, farm, or the like, acquiring various types of information and transmitting it to a server or the like. The charging element 20 outputs power to operate such a sensor. Since the power generation element 10, which converts environmental energy into power, cannot always generate stable power, the charging element 20 charges and stores the output power obtained by power generation by the power generation element 10 in order to stably operate the sensor.

[0015] The power supply module 30 includes a power supply device 50 and a control device 100 that controls the power supply device 50. The power supply device 50 includes a power supply circuit 51, an AD conversion unit 52, an AD conversion unit 53, and a DA conversion unit .

[0016] The power supply circuit 51 acquires the generated voltage from the power generation element 10 , performs processing such as boosting a voltage corresponding to the acquired generated voltage to generate output power, and outputs the generated output power to the charging element 20 .

[0017] The circuit configuration of the power supply module 30 according to the embodiment will be described with reference to Fig. 2. Fig. 2 is a circuit diagram of the power supply module according to the embodiment.

[0018] As shown in Figure 2, the circuit of the power generation element 10 is a simplified equivalent circuit of an electromagnetic induction type vibration power generation element, and is shown as an equivalent circuit in which a power source 11, a resistor R, and an inductor L are connected in series. The power source 11 outputs a generated voltage obtained by power generation using environmental energy. A positive electrode line PL is connected to the positive electrode side of the power source 11. A negative electrode line NL is connected to the negative electrode side of the power source 11. The resistor R and the inductor L are provided on the positive electrode line PL and connected in series to the power source 11.

[0019] The power supply circuit 51 includes switching elements S1 and S2 and diodes D1 and D2. In the example of Fig. 2, the power supply circuit 51 includes two switching elements S1 and S2, but the number of switching elements may be one, or three or more. The power supply circuit 51 is required to include at least one switching element.

[0020] The anode of the diode D1 is connected in series to the inductor L of the power generation element 10. The cathode of the diode D1 is connected to the charging element 20. The diode D2 is provided on the negative electrode line NL. The anode of the diode D2 is connected to the negative side of the power supply 11. The cathode of the diode D2 is connected between the cathode of the diode D1 and the charging element 20.

[0021] The switching elements S1 and S2 are configured, for example, by MOSFETs (Metal Oxide Semiconductor Field Effect Transistors). The drain of the switching element S1 is connected between the inductor L and the anode of the diode D1. The source of the switching element S1 is connected to ground. The gate of the switching element S1 is connected to the DA converter 54. The drain of the switching element S2 is connected between the negative electrode of the power supply 11 and the anode of the diode D2. The source of the switching element S2 is connected to ground. The gate of the switching element S2 is connected to the DA converter 54. Note that the switching elements S1 and S2 are not limited to MOSFETs and may be other semiconductor switches such as IGBTs (Insulated Gate Bipolar Transistors). The switching elements S1 and S2 are turned on / off by applying a voltage based on a control parameter from the control device 100 to their gates.

[0022] The power supply circuit 51 having such a configuration constitutes a diode rectification type boost converter together with the inductor L of the power generating element 10 .

[0023] The charging element 20 includes a charging capacitor C. One end of the capacitor C is connected to the cathode of the diode D1 and the cathode of the diode D2. The other end of the capacitor C is connected to ground.

[0024] The AD conversion unit 52 converts an analog signal output from the power supply device 50 into a digital signal and outputs the digital signal to the control device 100. For example, the AD conversion unit 52 acquires an analog signal corresponding to the generated voltage of the power generation element 10 from the power supply device 50, converts it into a digital signal, and outputs the digital signal corresponding to the generated voltage to the control device 100. The control device 100 can recognize the generated voltage of the power generation element 10 based on the digital signal corresponding to the generated voltage acquired via the AD conversion unit 52. One end of the AD conversion unit 52 is connected to the positive electrode line PL and negative electrode line NL on the input side of the power supply circuit 51. The other end of the AD conversion unit 52 is connected to the control device 100.

[0025] The AD conversion unit 53 converts an analog signal output from the power supply device 50 into a digital signal and outputs the digital signal to the control device 100. For example, the AD conversion unit 53 acquires an analog signal corresponding to the output power output from the power supply device 50, converts it into a digital signal, and outputs the digital signal corresponding to the output power to the control device 100. The control device 100 can recognize the output power output from the power supply device 50 based on the digital signal corresponding to the output power acquired via the AD conversion unit 53. One end of the AD conversion unit 53 is connected to the positive electrode line PL on the output side of the power supply circuit 51. The other end of the AD conversion unit 53 is connected to the control device 100.

[0026] The DA conversion unit 54 converts the digital signal output from the control device 100 into an analog signal and outputs the analog signal to the power supply circuit 51. For example, the DA conversion unit 54 acquires a digital signal corresponding to a control parameter output from the control device 100, converts it into an analog signal, and outputs the analog signal corresponding to the control parameter to the power supply circuit 51.

[0027] The control parameters are parameters for controlling the switching elements S1 and S2 of the power supply circuit 51. For example, the control parameters include the switching frequency, duty cycle, threshold voltage, delay time, and number of repetitions for the on / off operation of the switching elements S1 and S2, or data indicating the switching elements S1 and S2 whose on / off operation is to be controlled. For example, if the control device 100 determines the switching frequency as a control parameter, it outputs a digital signal of the voltage corresponding to the switching frequency to the DA conversion unit 54 so that the switching elements S1 and S2 perform on / off operation according to the switching frequency. The power supply circuit 51 performs on / off operation of the switching elements S1 and S2 based on the analog signal of the voltage corresponding to the control parameter obtained via the DA conversion unit 54. One end of the DA conversion unit 54 is connected to the gates of the respective switching elements S1 and S2. The other end of the DA conversion unit 54 is connected to the control device 100.

[0028] The control device 100 is configured to communicate with the power supply device 50, which includes a power supply circuit 51, and controls the power supply circuit 51 to generate output power based on the generated voltage of the power generation element 10. As shown in Figure 1, the control device 100 includes an arithmetic unit 101, a memory 102, a storage device 103, and a communication interface 104.

[0029] The arithmetic device 101 is a computing entity (computer) that executes predetermined processing. The arithmetic device 101 is configured with a processor such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), TPU (Tensor Processing Unit), or GPU (Graphics Processing Unit). A processor, which is an example of the arithmetic device 101, has the function of executing predetermined processing by executing a predetermined program. However, some or all of these functions may be implemented using dedicated hardware circuits such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array). The term "processor" is not limited to a processor in the narrow sense that executes processing using a stored program, such as a CPU, MPU, TPU, or GPU, but may also include hardwired circuits such as an ASIC or FPGA. The arithmetic device 101 described above can also be interpreted as a processing circuitry that executes predetermined processing. The arithmetic device 101 may be configured on a single chip or multiple chips. Furthermore, the processor and associated processing circuitry may be comprised of multiple computers interconnected by wire or wirelessly, such as via a local area network or a wireless network. The processor and associated processing circuitry may also be comprised of a cloud computer that performs remote calculations based on input data and outputs the results of the calculations to other devices at remote locations.

[0030] The memory 102 includes a volatile or non-volatile storage area (e.g., a working area) that temporarily stores program code, work memory, etc. 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.

[0031] The storage device 103 stores various programs executed by the arithmetic device 101, various data, and the like. For example, the storage device 103 stores an estimation program 131 executed by the arithmetic device 101 and an estimation model 132 used 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). The estimation program 131 and the estimation model 132 may be stored in the memory 102 (for example, a ROM).

[0032] The power supply module 30 configured as described above is used as a power supply for a wireless sensor node of an IoT (Internet of Things) device, etc. For this reason, the power supply module 30 is configured to efficiently extract output power from the generated voltage of the power generating element 10, output the output power to the charging element 20, and charge the charging element 20.

[0033] Specifically, the power generating element 10 generates power using environmental energy and outputs the generated voltage obtained by the power generation to the power supply circuit 51. The power supply circuit 51 switches the switching elements S1 and S2 between on and off in accordance with a predetermined duty ratio, thereby boosting the generated voltage obtained from the power generating element 10. At this time, the power supply circuit 51 performs the on / off operation of the switching elements S1 and S2 based on control parameters from the control device 100.

[0034] For example, the power supply circuit 51 performs the on / off operation of the switching elements S1, S2 in accordance with the switching frequency, duty ratio, threshold voltage, delay time, and number of repetitions in the on / off operation of the switching elements S1, S2, or the control target of the on / off operation, which are indicated by the control parameters from the control device 100. In this way, the control device 100 controls the on / off operation of the switching elements S1, S2 in the power supply circuit 51 based on the control parameters.

[0035] The power supply circuit 51 rectifies the voltage boosted by the on / off operation of the switching elements S1 and S2 using diodes D1 and D2. The power supply circuit 51 outputs output power corresponding to the voltage rectified by the diodes D1 and D2 to the charging element 20. The charging element 20 is charged using the output power obtained from the power supply circuit 51, and supplies stable power to the downstream sensor.

[0036] In this way, the control device 100 controls the power supply circuit 51 based on the control parameters to boost the generated voltage of the power generating element 10 and output output power corresponding to the required voltage to the charging element 20.

[0037] The power generating element 10 can be applied to a variety of power generating elements, such as photovoltaic power generating elements, thermoelectric power generating elements, electrostatic vibration power generating elements, and electromagnetic vibration power generating elements. The power supply circuit 51 must be developed individually for each combination of the power generating element 10 and the wireless sensor node. However, designing the power supply circuits 51 of the wireless sensor nodes individually to fit these various types of power generating elements 10 increases design costs. In particular, while the market for energy harvesting elements is small, there are a wide variety of power generation methods. Therefore, there are significant barriers to designing a power supply circuit 51 suitable for such small-lot, high-mix energy harvesting elements, making it difficult for the power supply circuits 51 for energy harvesting elements to become widespread.

[0038] Furthermore, if the power supply circuit 51 is individually designed to suit a specific power generating element 10, it is possible to configure a power supply circuit 51 that can efficiently obtain power for that specific power generating element 10, but the circuit configuration of the power supply circuit 51 becomes complex, and expandability such as the addition of other power generating elements 10 is lost. On the other hand, if the power supply circuit 51 is made versatile so that it can be adapted to various types of power generating elements 10, it becomes difficult to efficiently obtain power for each individual power generating element 10.

[0039] For this reason, there is a demand for a power supply module 30 that serves as a platform that can efficiently obtain power in accordance with various types of power generation elements 10, rather than a power supply circuit 51 that is specialized for a specific power generation element 10. Therefore, the power supply module 30 according to the embodiment is configured to dynamically determine optimal control parameters for maximizing the output power output from the power supply circuit 51, regardless of the type of power generation element 10 used.

[0040] Specifically, the control device 100 constructs a virtual power supply circuit 51 using the estimation model 132, and constructs a digital twin consisting of the actual power supply circuit 51 and a virtual power supply circuit corresponding to the power supply circuit 51. The control device 100 uses the digital twin of the power supply circuit 51 to estimate the output power output from the virtual power supply circuit based on the input parameters and control parameters of the power generating element 10 acquired from outside. The control device 100 then selects control parameters that maximize the output power based on the estimated value of the output power. More specifically, the control device 100 uses the virtual power supply circuit embodied by the estimation model 132, which is the digital twin of the power supply circuit 51, to estimate multiple output powers corresponding to each of multiple control parameters based on the input parameters of the power generating element 10 and the multiple control parameters, compares the estimated values ​​of the multiple output powers, and selects control parameters that maximize the estimated value of the output power.

[0041] The input parameters include the generated voltage or generated current of the power generating element 10. Furthermore, if the power generating element 10 is a thermoelectric power generating element, the input parameters include the temperature of the power generating element 10. If the power generating element 10 is an electrostatic vibration power generating element or an electromagnetic vibration power generating element, the input parameters include the vibration acceleration or vibration frequency of the power generating element 10. Thus, the input parameters include at least one of the generated voltage, generated current, temperature, vibration acceleration, and vibration frequency as parameters related to power generation by the power generating element 10.

[0042] This allows the user to efficiently obtain output power based on the generated voltage of the power generating element 10 through the boost operation of the power generating circuit 51, without having to design the power supply circuit 51 to match the power generating element 10, by simply outputting the control parameters selected by the control device 100 using the estimation model 132 to the power supply circuit 51.

[0043] FIG. 3 is a diagram illustrating an example of the estimation model 132. As shown in FIG. 3, the algorithm of the estimation model 132 is configured by an extreme learning machine. The extreme learning machine is a forward propagation neural network having only one hidden layer. Weights from the input layer to the hidden layer are not learned, but are generated randomly. In addition, the weights of the output layer are calculated using a pseudoinverse matrix or recursive least squares, which allows for faster learning than when using backpropagation.

[0044] FIG. 4 is a diagram illustrating another example of the estimation model 132. As shown in FIG. 4, the algorithm of the estimation model 132 may be configured with an echo state network. The echo state network is a type of reservoir computing model that processes time series data. The input layer receives the time series data and assigns appropriate weights. The reservoir layer uses an untrained recurrent neural network, creating a state in which past information from the time series data from the input layer reverberates and remains. Only the weights from the reservoir layer to the output layer are adjusted, and the features of the time series data are read out in the output layer. The echo state network enables faster learning than a typical recurrent neural network that learns all connection weights.

[0045] As shown in FIGS. 3 and 4 , during training of the estimation model 132, input parameters (e.g., power generation voltage) and control parameters (e.g., switching frequency) are used as input data for the estimation model 132, and output power is used as output data for the estimation model 132. When the input parameters and control parameters are input, the estimation model 132 estimates output power based on the input parameters and control parameters. During training, training data is prepared, each of which includes a set of the input parameters and control parameters and the output power actually obtained based on the input parameters and control parameters. The output power estimated by the estimation model 132 is compared with the output power included in the training data. If there is a difference between the estimated output power and the output power included in the training data, the estimation model 132 optimizes the algorithm to reduce the difference. In this way, the estimation model 132 learns the relationship between the input parameters, the control parameters, and the output power, thereby enabling it to accurately estimate the output power based on the input parameters and the control parameters.

[0046] Fig. 5 is a diagram showing simulation results related to training of the estimation model 132. The simulation shown in Fig. 5 shows changes in output power when training of the estimation model 132 is performed 2000 times, using the generated voltage as the input parameter and the switching frequencies of the switching elements S1 and S2 as the control parameters. Furthermore, in the simulation shown in Fig. 5, training of the estimation model 132 is performed using five different combinations of generated voltage and switching frequency. An extreme learning machine or an echo state network is used for the estimation model 132.

[0047] As shown in Figure 5, in either pattern, the output power converges to a constant value after approximately 900 repetitions of training the estimation model 132. Since one training session takes approximately one second, the training time required to achieve a constant output power is approximately 15 minutes. This training time is much faster than the training time required using general deep learning.

[0048] The power supply of the wireless sensor node to which the power supply module 30 is applied is required to be compact, since it is used for a sensor that operates standalone in a factory, farm, or the like. In this regard, it is easier to simplify the configuration of the estimation model 132 using an extreme learning machine or an echo state network than a learning machine that uses a deep neural network. Therefore, by applying an extreme learning machine or an echo state network to the estimation model 132, the power supply module 30 can be easily downsized.

[0049] The estimation model 132 may be trained in advance to estimate the output power output from the power supply circuit 51 in correspondence with the input parameters of the power generating element 10 and the control parameters of the power supply circuit 51. Furthermore, the estimation model 132 may be trained in real time while mounted on the power supply module 30 used as the power supply of the wireless sensor node so that each time an input parameter is input from the power generating element 10, the estimation model 132 estimates the output power actually output from the power supply circuit 51 in correspondence with the input parameter and the control parameter. That is, the estimation model 132 may be trained by learning not only in the training phase but also in the practical use phase.

[0050] The process executed by the control device 100 of the power supply module 30 according to the embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart showing the process executed by the control device 100. The process steps of the control device 100 shown in Fig. 6 are realized by the arithmetic device 101 executing the estimation program 131.

[0051] 6, the control device 100 determines (S1) whether or not the input parameters and control parameters have been acquired from the power supply circuit 51. If the control device 100 has not acquired the input parameters and control parameters from the power supply circuit 51 (NO in S1), the control device 100 ends this process.

[0052] On the other hand, if the control device 100 obtains input parameters and control parameters from the power supply circuit 51 (YES in S1), it uses a digital twin configured by the estimation model 132 to estimate the output power output from the virtual power supply circuit based on the input parameters and control parameters (S2). Based on the estimated values ​​of the control parameters, the control device 100 selects the control parameters that maximize the output power and outputs them to the power supply circuit 51 (S3), and then terminates this process.

[0053] As a result, the power supply circuit 51 can generate maximized output power using the generated voltage by operating the switching elements S1 and S2 based on the control parameters acquired from the control device 100.

[0054] In the example described above, the estimated model 132 was trained using the generated voltage as the input parameter and the switching frequencies of the switching elements S1 and S2 as the control parameters. However, other combinations of input and control parameters are also acceptable. For example, the input parameter can be any parameter related to the power generation of the power generation element 10, and at least one of the following may be used: generated voltage, generated current, temperature, vibration acceleration, and vibration frequency. The control parameter can be at least one of the following for at least one of the switching elements S1 and S2: switching frequency, duty cycle, threshold voltage, delay time, number of repetitions, and a control target to be selected from multiple options. Any combination of input and control parameters is acceptable as long as the control device 100 can use the estimated model 132 to estimate the control parameters for maximizing the output power output from the power supply circuit 51 based on the input parameters of the power generation element 10.

[0055] As described above, the control device 100 uses the digital twin of the power supply circuit 51 configured by the estimation model 132 to estimate the output power output from the virtual power supply circuit based on the input parameters and control parameters of the power generating elements 10, and can select optimal control parameters for maximizing the output power output from the power supply circuit 51. This makes it possible to select optimal control parameters for the power supply circuit 51 having the performance required to suit various types of power generating elements 10 by using the digital twin of the power supply circuit 51, without having to design the power supply circuit 51 individually to suit the power generating elements 10, and therefore it is possible to provide a power supply module 30 that can efficiently obtain output power to suit various types of power generating elements 10.

[0056] (Aspects) (1) A power supply module according to the present disclosure is a power supply module that outputs output power based on the generated voltage of a power generation element, and includes: a power supply circuit that generates output power using the generated voltage; and a control device that controls the power supply circuit based on control parameters, wherein the control device uses a digital twin of the power supply circuit configured by an estimation model to estimate output power output from a virtual power supply circuit based on input parameters and control parameters of the power generation element input to the power supply module, selects control parameters that maximize the output power based on the estimated value of the output power, and controls the power supply circuit based on the selected control parameters.

[0057] (2) In the power supply module described in (1) above, the estimation model is trained to estimate the output power actually output from the power supply circuit in accordance with the input parameters and control parameters each time an input parameter is input.

[0058] (3) In the power supply module described in (1) or (2) above, the power generation element includes an energy harvesting element that generates electricity using at least one of the following: light energy, thermal energy, vibration energy, flow energy, rotational energy, electromagnetic wave energy, electric field, or magnetic field energy.

[0059] (4) In the power supply module according to any one of (1) to (3) above, the input parameters include at least one of a generated voltage, a generated current, a temperature, a vibration acceleration, and a vibration frequency, which are related to power generation by the power generation element.

[0060] (5) In the power supply module described in any of (1) to (4) above, the power supply circuit includes at least one switching element that generates output power using the generated voltage based on control parameters, and the control parameters include at least one of the following for the switching element: switching frequency, duty cycle, threshold voltage, delay time, number of repetitions, and a control target to be selected from a plurality of options.

[0061] (6) In the power supply module according to any one of (1) to (5), the estimation model includes an extreme learning machine or an echo state network.

[0062] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims.

[0063] 1 Power generation system, 10 Power generation element, 11 Power supply, 20 Charging element, 30 Power module, 50 Power supply device, 51 Power circuit, 52, 53 AD conversion unit, 54 DA conversion unit, 100 Control device, 101 Arithmetic unit, 102 Memory, 103 Storage device, 104 Communication interface, 131 Estimation program, 132 Estimation model, C Capacitor, D1, D2 Diodes, L Inductor, NL Negative electrode wire, PL Positive electrode wire, R Resistor, S1, S2 Switching element.

Claims

1. A power supply module that outputs output power based on the generated voltage of a power generation element, comprising: a power supply circuit that generates the output power using the generated voltage; and a control device that controls the power supply circuit based on control parameters, wherein the control device uses a digital twin of the power supply circuit configured by an estimation model to estimate the output power output from a virtual power supply circuit based on the input parameters of the power generation element input to the power supply module and the control parameters, selects the control parameters that maximize the output power based on the estimated value of the output power, and controls the power supply circuit based on the selected control parameters.

2. The power supply module of claim 1, wherein the estimation model is trained to estimate the output power actually output from the power supply circuit in response to the input parameters and the control parameters each time the input parameters are input.

3. The power supply module according to claim 1 or claim 2, wherein the power generating element includes an environmental power generating element that generates power using at least one of light energy, thermal energy, vibration energy, flow energy, rotational energy, electromagnetic wave energy, electric field energy, and magnetic field energy.

4. A power supply module according to any one of claims 1 to 3, wherein the input parameters include at least one of the generated voltage, generated current, temperature, vibration acceleration, and vibration frequency, which are related to the power generation of the power generation element.

5. The power supply module according to any one of claims 1 to 4, wherein the power supply circuit includes at least one switching element that generates the output power using the generated voltage based on the control parameters, and the control parameters include at least one of the switching frequency, duty ratio, threshold voltage, delay time, number of repetitions, and a control target to be selected from a plurality of options for the at least one switching element.

6. A power supply module according to any one of claims 1 to 5, wherein the estimation model includes an extreme learning machine or an echo state network.

Images