Energy recovery and distribution method, device and equipment and readable storage medium

Abstract

The invention provides an energy recovery and distribution method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining an electric signal of an energy harvester under the condition that the vibration energy of a vehicle is collected through the energy harvester to generate electric energy, and converting the electric signal into a digital signal; inputting the digital signal into a pre-training model to obtain a predicted vibration frequency output by the pre-training model, and determining a current vibration frequency of the vehicle; according to the difference value between the predicted vibration frequency and the current vibration frequency, the capacitance of an impedance matching network is adjusted to adjust impedance; the impedance matching network is connected with the energy harvester; and distributing the electric energy generated by the energy harvester to at least one load. The energy transmission efficiency in the energy recovery and distribution process can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of vehicle technology, and in particular to an energy recovery and distribution method, device, equipment, and readable storage medium. Background Art

[0002] An energy harvester (EHR), also known as an energy harvester, is a device that converts idle energy in the environment (such as vibration, heat, light, wind, and electromagnetic energy) into electrical energy. It aims to solve the energy supply problem of low-power devices. Energy harvesters are passive, low-maintenance, and environmentally friendly, reducing dependence on batteries and lowering the deployment cost and environmental burden of electronic equipment.

[0003] In the prior art, when the electric energy generated by the energy harvester is used to supply the load, impedance mismatch often occurs, resulting in problems such as energy reflection and loss, power output reduction and poor stability, which significantly reduces the energy transmission efficiency. Summary of the Invention

[0004] In view of the above problems, embodiments of the present application are proposed to provide an energy recovery and distribution method, device, electronic device, and computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0005] In a first aspect, an embodiment of the present application discloses an energy recovery and distribution method, comprising:

[0006] In the case of collecting vibration energy of the vehicle through an energy harvester to generate electrical energy, obtaining an electrical signal from the energy harvester and converting the electrical signal into a digital signal;

[0007] Inputting the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model and determining a current vibration frequency of the vehicle;

[0008] According to the difference between the predicted vibration frequency and the current vibration frequency, adjusting the capacitance of the impedance matching network to adjust the impedance; the impedance matching network is connected to the energy harvester;

[0009] The electrical energy generated by the energy harvester is distributed to at least one load.

[0010] In a second aspect, an embodiment of the present application discloses an energy recovery and distribution device, comprising:

[0011] A signal conversion module, configured to obtain an electrical signal from an energy harvester when the energy harvester is used to harvest vibration energy of the vehicle to generate electrical energy, and convert the electrical signal into a digital signal;

[0012] a frequency prediction module, configured to input the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model and determine a current vibration frequency of the vehicle;

[0013] an impedance correction module, configured to adjust the capacitance of an impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency; the impedance matching network is connected to the energy harvester;

[0014] The electric energy distribution module is used to distribute the electric energy generated by the energy harvester to at least one load.

[0015] In a third aspect, an embodiment of the present application further discloses an electronic device comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the energy recovery and distribution method described in the first aspect are implemented.

[0016] In a fourth aspect, an embodiment of the present application further discloses a computer-readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the energy recovery and distribution method as described in the first aspect are implemented.

[0017] In an embodiment of the present application, when the vibration energy of a vehicle is collected by an energy harvester to generate electrical energy, an electrical signal of the energy harvester is obtained and converted into a digital signal, and the digital signal is input into a pre-trained model to obtain the predicted vibration frequency output by the pre-trained model, and the current vibration frequency of the generator is determined. According to the difference between the predicted vibration frequency and the current vibration frequency, the capacitance size of the impedance matching network is adjusted to adjust the impedance, and the electrical energy generated by the energy harvester is distributed to at least one load. When collecting the vibration energy of the vehicle, the vibration frequency can be predicted and compared with the actual current frequency, the impedance can be adjusted, and then the electrical energy can be distributed, thereby improving the efficiency of power transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a step diagram of an energy recovery and distribution method provided in an embodiment of the present application;

[0019] Figure 2 This is a step diagram of another energy recovery and distribution method provided by an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of the energy recovery and distribution process provided in an embodiment of the present application;

[0021] Figure 4 This is a schematic diagram of the energy recovery and distribution architecture provided in the embodiment of the present application.

[0022] Figure 5This is a block diagram of an energy recovery and distribution device provided in an embodiment of the present application;

[0023] Figure 6 This is a block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0025] Figure 1 A step diagram of an energy recovery and distribution method provided in an embodiment of the present application is shown, and the method includes:

[0026] Step 101 : When collecting vibration energy of a vehicle through an energy harvester to generate electrical energy, obtain an electrical signal from the energy harvester and convert the electrical signal into a digital signal.

[0027] In the embodiment of the present application, the energy harvester can generate electrical energy based on the vibration occurring in the vehicle by collecting vibration energy. The principle of the energy harvester is to convert mechanical energy such as vibration and kinetic energy into electrical energy. For example, it includes a piezoelectric energy harvester: using the piezoelectric effect of piezoelectric materials (such as quartz), the deformation of the material during vibration generates electric charge, which can be applied to industrial equipment vibration monitoring and wearable devices (such as walking power insoles); an electromagnetic energy harvester: generating induced electromotive force by cutting magnetic flux lines or permanent magnet movement through a coil, which can be applied to small wind turbines and rotating machinery (such as bicycle hub power generation); an electrostatic energy harvester: based on the change in the spacing or area of ​​parallel plate capacitors, it generates electricity through charge accumulation and can be applied to low-frequency vibration scenarios (such as human body movement). The energy harvester in the present application can be a composite energy harvester, such as a piezoelectric-electromagnetic composite energy harvester, which can achieve broadband energy capture. Compared with traditional single energy capture technology, the energy capture efficiency is significantly improved, the full-band vibration energy of the generator is effectively utilized, energy waste is reduced, and dependence on external power supply is reduced, meeting the energy needs of the self-powered monitoring system.

[0028] The vibration energy of a vehicle can be generated at different locations of the vehicle, mainly concentrated in locations that can generate vibration, pressure or deformation, such as the generator, engine, suspension system, hub cover, wheel hub or inside the tire. The location where the vehicle vibration energy is generated is not specifically limited here.

[0029] The high-speed analog-to-digital converter (ADC) module built into the field programmable gate array (FPGA) boasts nanosecond sampling speeds, matching the dynamic output characteristics of the piezoelectric-electromagnetic composite energy harvester (e.g., microsecond-level voltage fluctuations generated by high-frequency vibrations). Based on the frequency range of the energy harvester's output signal (e.g., the vibration frequency of piezoelectric materials typically ranges from tens to hundreds of hertz), and in accordance with the Nyquist sampling theorem, the ADC can be set at a sufficiently high sampling rate, such as at least twice the signal's highest frequency, to avoid aliasing distortion.

[0030] The analog electrical signals output by the energy harvester, such as the charge signal generated by the vibration of the piezoelectric material and the induced electromotive force caused by the electromagnetic coil cutting the magnetic flux lines, are usually weak signals and are mixed with environmental noise (such as electromagnetic interference from vehicle vibration). Therefore, high-frequency noise such as switching power supply ripple and radio frequency interference can be filtered out through digital filters or analog front-end filtering circuits, and components above the signal frequency band can be attenuated through low-pass filters. If periodic interference (such as motor power frequency noise) exists in the energy harvester's operating environment, it can be specifically suppressed through band-stop filters (notch filters).

[0031] In addition, since the original signal amplitude output by the energy harvester is usually small, the signal amplitude can be increased through a programmable gain amplifier or a signal amplification circuit.

[0032] The filtered and amplified analog signal can be discretized into a binary digital signal (such as 12-bit or 16-bit accuracy) through architectures such as successive approximation registers and pipelines. The digitized signal is transmitted in parallel or serially to the internal logic unit of the FPGA for subsequent processing (such as energy management, signal analysis, storage, etc.).

[0033] Step 102 : Input the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model, and determine the current vibration frequency of the vehicle.

[0034] The pre-trained model can be a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory network (LSTM), and other models.

[0035] First, an energy harvester can be used to collect electrical signals generated by vehicle vibrations, while simultaneously recording the actual vibration frequency as a label. The collected electrical signals are preprocessed through filtering, noise reduction, and feature extraction. Data augmentation can also be used to expand the sample size. Models can include CNNs, RNNs, or hybrid models. During training, regression loss functions such as mean squared error (MSE) are used, along with optimizers such as Adam, combined with regularization and early stopping to prevent overfitting. After training, model performance is evaluated using mean squared error and mean absolute percentage error (MAPE). If the model performs poorly, optimization methods such as feature fusion and ensemble learning can be attempted. Finally, the trained model is quantized and deployed on an FPGA. New data can also be used for regular incremental training. The method for obtaining the pretrained model is not specifically limited here.

[0036] When a digital signal is input into a pre-trained model, such as a neural network model, its cell state and gating mechanisms (forget gate, input gate, and output gate) are used to process the input data. The forget gate determines which historical information is discarded from the cell state; the input gate filters and updates the new information required by the cell state; and the output gate generates output based on the updated cell state. This allows the model to extract key features from historical frequency data and memorize important information related to frequency change trends. Based on learning and feature extraction from historical data, the pre-trained model can predict the vibration frequency change trend over a period of time (e.g., 10-50 milliseconds) and output the predicted frequency value.

[0037] While obtaining the predicted vibration frequency output by the pre-trained model, the vibration frequency of the component that generates vibration energy in the vehicle, such as the generator, can be monitored in real time through sensors to obtain the current vibration frequency.

[0038] Step 103: adjusting the capacitance of an impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency; the impedance matching network is connected to the energy harvester.

[0039] As the core control unit of the system, the FPGA can receive the predicted frequency value output by the LSTM neural network in real time and compare it with the current real-time frequency value collected by the sensor. After calculating the difference between the two through subtraction or other operations, such as the difference, the difference is compared and analyzed with a preset frequency change threshold (such as 5 Hz). When the predicted frequency is higher than the current frequency and the difference exceeds the threshold, it indicates that the vibration frequency is on an upward trend. At this time, the FPGA can determine that the impedance needs to be reduced to achieve impedance matching. Conversely, if the predicted frequency is lower than the current frequency and the difference exceeds the limit, the impedance needs to be increased to achieve impedance matching.

[0040] Among them, the impedance matching network can be an LC matching network, L stands for inductor, which is a component that can convert electrical energy into magnetic energy and store it. It is usually made of a wound wire (such as a hollow coil, an iron core coil, etc.). Its characteristic is that it has an obstruction effect on alternating current (inductive reactance), and the size of the obstruction is related to the inductance value and the current frequency; C stands for capacitor, which is a component that can store electric charge. It is composed of two conductive plates and an insulating medium in between. Its characteristic is that it has a conducting effect on alternating current (capacitive reactance), and the size of the capacitive reactance is inversely proportional to the capacitance value and the current frequency.

[0041] An impedance matching network is a circuit network composed of inductors and capacitors, used for impedance matching. That is, in a signal transmission or energy conversion system, by adjusting the parameters of the inductors and capacitors, the output impedance of the power supply (or signal source) is matched with the load impedance, thereby maximizing energy transmission and optimizing system performance.

[0042] Impedance matching networks can have various topologies, such as the L-type network: consisting of an inductor and a capacitor, with a simple structure, used for single-port impedance matching; the π-type network: consisting of three elements (such as two capacitors and an inductor), which can achieve more flexible impedance transformation; and the T-type network: consisting of three elements (such as two inductors and a capacitor), which is suitable for scenarios requiring wideband matching.

[0043] During the vibration energy harvesting process, inductors and capacitors of different specifications can be dynamically switched, such as through metal oxide semiconductor field effect transistor (MOSFET) switches. After determining the adjustment strategy, the FPGA generates the corresponding control signal through the internal logic circuit. The control signal is transmitted to the drive circuit in the form of an electrical signal to control the on and off state of the MOSFET switch. For example, when the inductance value needs to be reduced, the FPGA can send a specifically coded control signal. After amplification by the drive circuit, it turns on the MOSFET switch corresponding to the smaller inductance value, and then connects the inductor to the impedance matching network, so that the impedance matching network parameters adapt to the changes in the vibration frequency in real time, thereby maintaining efficient energy conversion.

[0044] Step 104: Distribute the electrical energy generated by the energy harvester to at least one load.

[0045] The electrical energy generated by the energy harvester can be used to supply at least one load, and the distribution can be to power each load in sequence according to a fixed cycle (for example, every 100 milliseconds, the sensor and communication module are alternately powered), which is suitable for low-power, non-real-time load acquisition. Alternatively, dynamic priority distribution can be used to dynamically adjust the power supply sequence based on the load status. For example, when the vibration frequency is abnormal, the fault diagnosis sensor is given priority, and when the wireless communication module needs to transmit data, its power supply priority is temporarily increased. Alternatively, proportional distribution can be used to distribute energy to multiple loads in a fixed ratio through a resistor divider or current mirror circuit. There is no specific limitation on how the electrical energy generated by the energy harvester is distributed to the at least one load.

[0046] In summary, in the embodiment of the present application, when the vibration energy of the vehicle is collected by an energy harvester to generate electrical energy, the electrical signal of the energy harvester is obtained and converted into a digital signal, and the digital signal is input into a pre-trained model to obtain the predicted vibration frequency output by the pre-trained model, and the current vibration frequency of the generator is determined. According to the difference between the predicted vibration frequency and the current vibration frequency, the capacitance size of the impedance matching network is adjusted to adjust the impedance, and the electrical energy generated by the energy harvester is distributed to at least one load. When collecting the vibration energy of the vehicle, the vibration frequency can be predicted and compared with the actual current frequency, the impedance can be adjusted, and then the electrical energy can be distributed, thereby improving the efficiency of power transmission.

[0047] refer to Figure 2 , which shows a flow chart of the steps of an energy recovery and distribution method provided by an embodiment of the present application, the method comprising:

[0048] Step 201: When a vibration energy of a vehicle is collected by an energy harvester to generate electrical energy, an electrical signal of the energy harvester is obtained, and the electrical signal is converted into a digital signal;

[0049] Step 202: input the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model, and determine the current vibration frequency of the vehicle;

[0050] Step 203: adjusting the capacitance of an impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency; the impedance matching network is connected to the energy harvester;

[0051] Step 204: Distribute the electrical energy generated by the energy harvester to at least one load.

[0052] The above steps 201 to 204 can refer to the above Figure 1 The contents of the embodiments will not be repeated here.

[0053] Optionally, after step 203 of adjusting the capacitance of the impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency, the method further includes:

[0054] Step A1, determining output power, load power, output impedance, and load impedance based on voltages and currents corresponding to the energy harvester and each load, and determining current power transmission efficiency based on the output power and the input power;

[0055] Step A2: comparing the current power transmission efficiency with the historical power transmission efficiency, and comparing the output impedance with the load impedance;

[0056] Step A3: When the current power transmission efficiency is less than the historical power transmission efficiency, or the output impedance and the load impedance do not match, the current power transmission efficiency, the output impedance and the load impedance are input into the pre-training model for training.

[0057] In the embodiment of the present application, after the impedance matching network adjustment is completed, the voltage, current and other parameters in the circuit can be continuously monitored, and the effect of the parameter adjustment can be evaluated by calculating the energy transmission efficiency and real-time impedance. If it is found that the energy transmission efficiency is improved and the impedance matching is good after the adjustment, it means that the adjustment strategy is correct; if the efficiency is not improved or impedance mismatch occurs, it indicates that there is a problem with the current adjustment strategy.

[0058] The output power and output impedance of the energy harvester can be calculated based on the voltage and current corresponding to the energy harvester, and the load power and load impedance of each load can be calculated based on the voltage and current corresponding to each load, where the impedance of each component can be calculated by dividing the voltage by the current.

[0059] The current power transmission efficiency is compared with the historical power transmission efficiency to determine whether the power transmission efficiency has increased or decreased. The historical power transmission efficiency can be multiple power transmission efficiencies before the impedance matching network is adjusted. Similarly, each historical power transmission efficiency can be calculated based on the voltage and current corresponding to each component.

[0060] It is also necessary to compare the output impedance and the load impedance. Since the purpose of adjusting the impedance matching network is to achieve conjugate matching and thus improve the efficiency of power transmission, it is necessary to compare the output impedance and the load impedance after adjustment of the impedance matching network to determine whether conjugate matching has been achieved.

[0061] If the current power transmission efficiency is lower than the historical power transmission efficiency, or if the output impedance and load impedance do not match, indicating poor adjustment results, new feedback data, such as the current power transmission efficiency, output impedance, load impedance, and actual vibration frequency, can be added to the neural network model's training sequence to trigger model fine-tuning. The neural network model relearns the relationship between frequency and impedance under the current operating conditions based on the new data, optimizing the prediction model and providing a more accurate basis for the next prediction and control. The above detection and adjustment process is continuously looped during the operation of the generator, continuously collecting data, predicting frequency trends, adjusting the impedance matching network, and implementing real-time adaptive adjustment of the generator's vibration frequency changes based on the feedback optimization strategy, ensuring that the energy recovery system always maintains an efficient operating state.

[0062] In an embodiment of the present application, the output power, load power, output impedance and load impedance are determined based on the voltage and current corresponding to the energy harvester and each load respectively, and the current power transmission efficiency is determined based on the output power and the input power. The current power transmission efficiency is compared with the historical power transmission efficiency, and the output impedance is compared with the load impedance. When the current power transmission efficiency is lower than the historical power transmission efficiency, or the output impedance and the load impedance do not match, the current power transmission efficiency, output impedance and load impedance are input into the pre-trained model for training. This adaptive training mechanism based on multi-parameter monitoring and comparison can improve the energy efficiency and stability of the energy harvesting system. By calculating the output power, load power and transmission efficiency in real time, the system can quantitatively evaluate the current energy transmission status, identify the performance degradation trend through historical efficiency comparison, trigger optimization in time, and restore system performance.

[0063] Optionally, the step 204 of distributing the electric energy generated by the energy harvester to at least one load includes:

[0064] Step 2041: When the electric energy generated by the energy harvester is greater than the electric energy consumption of at least one load, the electric energy generated by the energy harvester is distributed to at least one load, and the electric energy generated by the energy harvester is stored in an energy storage component.

[0065] Step 2042: When the electric energy generated by the energy harvester is less than the electric energy consumption of at least one load, distribute the electric energy generated by the energy harvester and the electric energy in the energy storage assembly to at least one load.

[0066] In the embodiments of the present application, the output power of the energy harvester and the total power of the load can be monitored in real time using sensors such as voltage and current sensors. When the power of the energy harvester exceeds the load demand, the power can be stepped down and stabilized by a DC-DC converter (Direct Current-Direct Current converter). A portion of the power is then distributed to the load via a multi-way switch, while the remaining portion is stored in a lithium battery or supercapacitor via a charge management chip. When the power of the energy harvester is insufficient, the energy storage component can be activated, and the DC-DC boost circuit combines the stored power with the power of the energy harvester, dynamically adjusting the output voltage to meet the load demand.

[0067] For example, in a vehicle vibration energy harvesting system, the energy harvester generates 200 milliwatts (mW) of electricity when the vehicle is bumpy, while the total demand of the sensor (50mW) and the communication module (100mW) is 150mW. The excess 50mW of electricity will be stored in the supercapacitor; when the vehicle is driving smoothly, the energy harvester only outputs 30mW. At this time, the energy storage component can cooperate with the energy harvester to provide power, giving priority to ensuring the operation of the sensor, and the communication module enters low-power mode.

[0068] By implementing the embodiments of the present application, when the electric energy generated by the energy harvester is greater than the electric energy consumption of at least one load, the electric energy generated by the energy harvester is distributed to at least one load, and the electric energy generated by the energy harvester is stored in the energy storage component. When the electric energy generated by the energy harvester is less than the electric energy consumption of at least one load, the electric energy generated by the energy harvester and the electric energy in the energy storage component are distributed to at least one load. This can achieve dynamic energy balance, store electric energy when there is surplus energy to avoid waste, enable energy storage when energy is insufficient, ensure uninterrupted operation of critical loads, improve system endurance, and at the same time effectively improve system reliability by prioritizing power supply to critical loads.

[0069] Optionally, the energy storage component includes a first energy storage component and a second energy storage component; the step 2041 of storing the electric energy generated by the energy harvester in the energy storage component includes:

[0070] Step 20411: Based on a first preset charge threshold and a first preset voltage threshold of the first energy storage component, store the electrical energy generated by the energy harvester in the first energy storage component until the first energy storage component is fully charged; the first preset charge threshold is used to trigger charging of the first energy storage component with a first preset current; the first preset voltage threshold is used to trigger charging of the first energy storage component with a first preset voltage;

[0071] Step 20412: When the first energy storage component is fully charged, based on a second preset charge threshold and a second preset voltage threshold of the second energy storage component, the electric energy generated by the energy harvester is stored in the second energy storage component until the second energy storage component is fully charged; the second preset charge threshold is used to trigger charging of the second energy storage component with a second preset current; the second preset voltage threshold is used to trigger charging of the second energy storage component with a second preset voltage.

[0072] In an embodiment of the present application, when the generator waste vibration energy recovery system is started, the energy storage state maintenance module is initialized synchronously, the energy storage unit management module is started and establishes a communication connection with each component in the hierarchically managed hybrid energy storage system, reads the initial parameters of each energy storage component, such as the initial voltage of the supercapacitor, the charge state of the lithium-ion capacitor, the initial speed of the micro flywheel, etc., and records these data in the system database. At the same time, various sensors distributed in the energy storage system (voltage sensor, current sensor, temperature sensor, flywheel speed sensor, etc.) begin to collect data in real time and transmit the collected raw data to the energy storage unit management module:

[0073] The first and second energy storage components in a hybrid energy storage system can be the same or different. The first energy storage component can be a graphene composite supercapacitor, serving as the primary energy storage unit, while the second energy storage component can be a lithium-ion capacitor-micro-flywheel energy storage combination, serving as a transient buffer unit. The graphene composite supercapacitor uses graphene as its electrode material, leveraging its high surface area and excellent conductivity to achieve rapid charge and discharge speeds and high power density. Compared to traditional capacitors, it can store more energy and has a long cycle life, with charge and discharge cycles reaching hundreds of thousands of times. It is suitable for scenarios such as rapid start-stop in electric vehicles and peak shaving in smart grids. The lithium-ion capacitor-micro-flywheel energy storage combination combines the high energy density of the lithium-ion capacitor with the high power density and long life of the micro-flywheel. The lithium-ion capacitor provides basic energy storage, while the micro-flywheel rapidly releases energy when needed and quickly stores excess energy when it is in excess. Working together, the two combine energy storage capacity and power response speed, enabling applications such as instantaneous power supply for devices and uninterruptible power supply for high-end servers, improving system reliability and efficiency. The specific types of the first energy storage assembly and the second energy storage assembly are not particularly limited herein.

[0074] The charging process of the first energy storage component and the second energy storage component can adopt a staged control strategy: the real-time state of charge of the first energy storage component is compared with a first preset charge threshold. If the real-time state of charge is equal to the first preset charge threshold, the state of charge of the first energy storage component is lower, and a constant current charging method can be adopted to quickly charge with a larger current to improve charging efficiency; when the real-time voltage of the first energy storage component reaches the first preset voltage threshold, indicating that the first energy storage component is close to the full charge voltage, the constant voltage charging method based on the first preset voltage is switched to, thereby performing trickle charging with a smaller current to ensure that the first energy storage component is fully charged and not overcharged.

[0075] When the first energy storage component is fully charged, the second energy storage component can be charged. Similarly, the real-time state of charge of the second energy storage component is compared with the second preset charge threshold. If the real-time state of charge is equal to the second preset charge threshold, the second energy storage component has a lower state of charge, and a constant current charging method can be used to quickly charge with a larger current to improve charging efficiency. When the real-time voltage of the second energy storage component reaches the second preset voltage threshold, indicating that the second energy storage component is close to the full charge voltage, the constant voltage charging method based on the second preset voltage is switched to, thereby performing trickle charging with a smaller current to ensure that the second energy storage component is fully charged and not overcharged.

[0076] During the charging process of each energy storage component, the respective states of the first energy storage component can be monitored, for example, by continuously monitoring the terminal voltage of the graphene composite supercapacitor through a voltage sensor, detecting the charge and discharge current in real time through a current sensor, and monitoring its operating temperature through a temperature sensor. Parameters such as the state of charge and state of health (SOH) can also be calculated. If an abnormal situation occurs, such as excessive temperature or abnormal fluctuations in charging current, the energy storage unit management module used to manage the energy storage component can immediately suspend charging, start a fault diagnosis program, analyze the cause of the abnormality, and take protective measures, such as reducing the charging current, starting a heat dissipation device, etc. A charging protection mechanism can also be set for each energy storage component, for example, a voltage sensor monitors the voltage across the energy storage component in real time. When the detected value approaches a preset full charge threshold, a signal is transmitted to a control chip such as a microcontroller unit (MCU) or a dedicated protection integrated circuit (IC). The control chip triggers a logic signal to drive an electronic switch (such as a MOSFET) to cut off the charging circuit and prevent current from continuing to flow in. The charging protection mechanism can also be implemented through shunt resistors or energy-consuming elements. By connecting shunt resistors in parallel, when the voltage exceeds the threshold, the excess energy is released through active energy consumption (such as resistor heating) to achieve soft protection.

[0077] Alternatively, a low-voltage warning can be activated when the voltage drops below a set minimum operating voltage threshold, alerting the system to a potential energy shortage. For example, a flywheel speed sensor monitors the flywheel's speed in real time and calculates its energy storage status based on the speed. If the speed is too high, approaching a safety threshold, the brakes are activated to prevent damage. If the speed drops too low, affecting instantaneous power output, the energy storage unit management module can replenish the flywheel's energy storage.

[0078] When the system encounters a transient high-power demand (such as a large transient load), the transient buffer unit can respond quickly, and the lithium-ion capacitor and micro-flywheel energy storage work together to release energy. The micro-flywheel converts the stored kinetic energy into electrical output by rapidly reducing its speed; the lithium-ion capacitor discharges with a large current to meet the transient high-power demand. The energy storage unit management module can monitor various parameters during the discharge process in real time. When the lithium-ion capacitor voltage drops to the set minimum discharge voltage or the micro-flywheel speed drops to the minimum operating speed, the protection mechanism is activated to stop discharge and prevent excessive discharge from damaging the energy storage components.

[0079] In the implementation of the embodiment of the present application, based on the first preset charge threshold and the first preset voltage threshold of the first energy storage component, the electric energy generated by the energy harvester is stored in the first energy storage component until the first energy storage component is fully charged. When the first energy storage component is fully charged, based on the second preset charge threshold and the second preset voltage threshold of the second energy storage component, the electric energy generated by the energy harvester is stored in the second energy storage component until the second energy storage component is fully charged. From the perspective of charge and discharge life, based on the preset charge threshold and voltage threshold, a staged and stepped charging method is adopted to avoid irreversible damage to the energy storage component caused by direct charging with high current and high voltage, extend the cycle life of the first and second energy storage components, and reduce replacement costs. In terms of charging efficiency, different currents and voltages are matched at different stages. When the energy storage component is low in power, high current can be used to quickly replenish electric energy. When it is close to full charge, a stable voltage can be used to prevent overcharging, taking into account both speed and safety, and improving energy storage efficiency as a whole. In terms of system stability, graded charging enables energy storage components to operate under appropriate parameters, reducing the risks of overheating and bulging caused by improper charging, ensuring continuous and stable operation of the system. At the same time, it reasonably allocates the order of power storage, optimizes energy management, and enhances the system's ability to cope with complex working conditions.

[0080] During the operation of each energy storage component, health status assessment and maintenance can be performed. The energy storage unit management module regularly (e.g., once a day) assesses the health status of each energy storage component of the hybrid energy storage system. For graphene composite supercapacitors, the SOH is calculated by analyzing parameters such as their charge and discharge efficiency, internal resistance change, and voltage holding capacity, combined with relevant health assessment models. When the SOH is below 80%, a maintenance reminder can be issued, recommending inspection and maintenance of the supercapacitor; when the SOH is below 60%, a reminder to replace the supercapacitor is issued to ensure the system's energy storage performance.

[0081] For the lithium-ion capacitor-micro flywheel energy storage combination, health assessments are performed on the lithium-ion capacitor and micro flywheel separately. The lithium-ion capacitor's state of health is assessed based on parameters such as the number of charge and discharge cycles, capacity decay rate, and internal resistance change; the micro flywheel's health is assessed by monitoring indicators such as bearing wear, rotor dynamic balance, and speed stability. When the assessment results indicate a potential failure risk, the energy storage unit management module promptly issues an early warning and provides corresponding maintenance recommendations, such as balanced charging of the lithium-ion capacitor and dynamic balancing calibration of the micro flywheel.

[0082] Data recording and optimization can also be performed. The energy storage unit management module records all key data in the energy storage status maintenance process, including the voltage, current, temperature, SOH, charge and discharge times, charge and discharge power, etc. of each energy storage component, in real time to the system database. By analyzing historical data, the potential patterns in the data are mined to provide data support for optimizing energy storage strategies and improving energy storage system design. For example, by analyzing the charge and discharge curves of supercapacitors, the charging strategy can be adjusted to improve charging efficiency; by comparing the capacity attenuation of lithium-ion capacitors in different time periods, their service life can be evaluated and replacement preparations can be made in advance. At the same time, these data can also be used to communicate with the host computer to provide operators with operating status information of the energy storage system, facilitating remote monitoring and management.

[0083] Collected generator operating data (such as vibration, temperature, and current) and energy recovery and power supply system status data (such as energy harvester output power and energy storage unit charge) can also be transmitted to the host computer in real time. Through the host computer's monitoring software, operators can intuitively view the generator's operating status and the system's energy status. The host computer stores and analyzes historical data, using data analysis algorithms to uncover underlying patterns within the data, providing data support for subsequent power generation system optimization and energy management strategy adjustments. For example, by analyzing long-term vibration data, potential generator failure hazards can be predicted; based on energy recovery data, the installation location and parameter settings of the composite energy harvester can be optimized. Furthermore, operators can send control commands to the system through the host computer to manually intervene in the monitoring system's operating mode, such as adjusting the sensor sampling frequency or energy distribution strategy.

[0084] Optionally, the load is a sensor, and each sensor corresponds to its own detection threshold; the step 204 of distributing the electric energy generated by the energy harvester to at least one load includes:

[0085] Step 2043, determining the parameter value detected by each sensor, and comparing the parameter value with the detection threshold corresponding to the sensor;

[0086] Step 2044: If the parameter value is greater than the detection threshold, the sensor is determined to be in an abnormal state; or if the parameter value is less than or equal to the detection threshold, the sensor is determined to be in a normal state, and then a state combination corresponding to at least one sensor is determined;

[0087] Step 2045 : Determine the distribution ratio corresponding to the state combination in the pre-built mapping relationship, and distribute the electric energy generated by the energy harvester to at least one sensor based on the distribution ratio; the distribution ratio represents the proportion of electric energy corresponding to each sensor.

[0088] In an embodiment of the present application, the load may be a sensor, and each sensor may have a corresponding detection threshold. Since the sensor may be used to detect various parameters, such as temperature, speed, etc., the parameters detected by the sensor usually have a range, namely the detection threshold.

[0089] The parameter value detected by each sensor is determined and compared with the sensor's corresponding detection threshold. If the parameter value is greater than the detection threshold, it indicates that the parameter detected by the sensor is outside the normal range, and the sensor is determined to be in an abnormal state. Similarly, if the parameter value is less than or equal to the detection threshold, the sensor is determined to be in a normal state.

[0090] After determining the state of each sensor, the state combination corresponding to at least one sensor can be obtained. For example, three sensors are configured to monitor current, voltage, and temperature, respectively. The detection thresholds, or significance thresholds, are: current threshold = 10, voltage threshold = 20, and temperature threshold = 30. The parameters D detected by the three sensors are [15, 10, 5] (current = 15, voltage = 10, temperature = 5). Comparing the sensor parameter values ​​with the thresholds generates binary states: current 15 > 10 → state bit 1 (indicating an abnormal state), voltage 10 < 20 → state bit 0 (indicating a normal state), and temperature 5 < 30 → state bit 0. The resulting state combination can be the binary state S = [1, 0, 0], which can be further converted to decimal S = 4 (i.e., 2² = 4).

[0091] The pre-established mapping relationship can be obtained through testing and calibration. The pre-established mapping relationship can include the relationship between state combinations and allocation ratios. Therefore, the allocation ratio corresponding to each state combination can be determined in the pre-established mapping relationship. Based on the allocation ratio, the electrical energy generated by the energy harvester is allocated to at least one sensor. For example, the determined allocation ratio may be a = [6, 2, 2] (i.e., 60% to the current sensor, 20% to the voltage sensor, and 20% to the temperature sensor).

[0092] In implementing the embodiments of the present application, by determining the parameter value detected by each sensor and comparing the parameter value with the detection threshold corresponding to the sensor, when the parameter value is greater than the detection threshold, the sensor is determined to be in an abnormal state, and when the parameter value is less than or equal to the detection threshold, the sensor is determined to be in a normal state, and then the state combination corresponding to at least one sensor is determined, and in the pre-constructed mapping relationship, the distribution ratio corresponding to the state combination is determined, and based on the distribution ratio, the electric energy generated by the energy harvester is distributed to at least one sensor. From the perspective of equipment maintenance, real-time comparison of parameter values ​​and detection thresholds can quickly locate abnormal sensors, avoid system failures caused by sensor failures, and reduce maintenance costs and downtime risks. In terms of power utilization efficiency, the power distribution ratio is adjusted according to the sensor status, and the power supply of abnormal sensors is prioritized to avoid power waste, ensure the energy supply of key monitoring tasks, and improve overall energy utilization.

[0093] Optionally, the mapping relationship includes a first mapping relationship and a second mapping relationship; the step 2045 of determining the allocation ratio corresponding to the state combination in the pre-constructed mapping relationship includes:

[0094] Step 20451: Determine, in the first mapping relationship, at least one allocation ratio corresponding to the state combination, and determine, in the second mapping relationship, at least one allocation ratio corresponding to the state combination; each allocation ratio is associated with an index and a value parameter; the same allocation ratio in different mapping relationships is associated with the same index and different value parameters;

[0095] Step 20452: for each index, determine the value parameter mean corresponding to the index to obtain at least one value parameter mean;

[0096] Step 20453: Determine the maximum value parameter mean among at least one value parameter mean, and determine the target index corresponding to the maximum value parameter mean, and the target allocation ratio corresponding to the target index.

[0097] In an embodiment of the present application, the mapping relationship includes a first mapping relationship and a second mapping relationship, each of which includes a correspondence between multiple allocation ratios, indexes, and value parameters. Among them, the mapping relationship can be obtained based on the double Q learning algorithm, and two Q tables can be maintained, namely the Q1 table and the Q2 table. During initialization, the Q values ​​in these two tables are initialized to small random values ​​or 0. During the training process, for each state (corresponding to the normal state or abnormal state of the sensor) and action (corresponding to the allocation ratio, the allocation ratio represents the action of allocating electrical energy to different sensors), the action will be selected according to the current strategy (such as the greedy strategy), and the Q table will be updated according to the reward and the next state of the environmental feedback. When updating, the Q1 table or the Q2 table will be selected for updating with a certain probability (such as 0.5). For example, if the Q1 table is chosen to be updated, then the Q2 table is used to estimate the target Q value according to the Q learning update rule, and then the Q value (i.e., the value parameter) of the corresponding state-action pair in the Q1 table is updated, and vice versa. In this way, the two Q-tables are independent of each other and cooperate with each other, gradually learning the optimal state-action value function, and then obtaining the Q-table, that is, the mapping relationship.

[0098] In the first mapping relationship, at least one allocation ratio corresponding to the state combination is determined, and in the second mapping relationship, at least one allocation ratio corresponding to the state combination is determined. For example, in the first mapping relationship, the at least one allocation ratio determined can be a-1, a-2, a-3 (1, 2, 3 are indexes), and the corresponding value parameters can be 2.5, 3.0, and 1.8, respectively. Since the same allocation ratio in different mapping relationships corresponds to the same index in the first mapping relationship, in the second mapping relationship, the at least one allocation ratio determined can also be a-1, a-2, a-3, but the value parameters can be 2.9, 2.7, and 2.1, respectively. For each index, the mean value parameter corresponding to the index is determined. The mean value parameter corresponding to index 1 is (2.5+2.9)÷2=2.7, the mean value parameter corresponding to index 2 is (3.0+2.7)÷2=2.85, and the mean value parameter corresponding to index 3 is (1.8+2.1)÷2=1.95. Then the target index is 2, and the allocation ratio corresponding to the target index is the final target allocation ratio. Electric energy can be allocated based on the target allocation ratio.

[0099] For example, the allocation ratio corresponding to index 2 is a = [6, 2, 2] (i.e., 60% for the current sensor, 20% for the voltage sensor, and 20% for the temperature sensor). Based on this, the actual energy distribution is calculated. The current energy generated by the energy harvester (or the energy provided by the energy storage component, or the energy provided by the energy harvester and the energy storage component) can be discretized and graded. For example, 10 levels (from 1 to 10) can be set. The current energy levels are combined and the allocation ratios are converted into actual energy values. For example, if the current sensor E1 = 8 × (6 / 10) = 4.8, the voltage sensor E2 = 8 × (2 / 10) = 1.6, and the temperature sensor E3 = 8 × (2 / 10) = 1.6, the final energy distribution scheme is [4.8, 1.6, 1.6].

[0100] In an embodiment of the present application, at least one allocation ratio corresponding to a state combination is determined in a first mapping relationship, and at least one allocation ratio corresponding to a state combination is determined in a second mapping relationship; each allocation ratio is associated with an index and a value parameter; the same allocation ratio is associated with the same index and different value parameters in different mapping relationships;

[0101] For each index, the mean of the value parameters corresponding to the index is determined to obtain at least one value parameter mean. The maximum value parameter mean among the at least one value parameter mean is determined, and the target index corresponding to the maximum value parameter mean and the target allocation ratio corresponding to the target index are determined. Taking the mean value can integrate the judgments of the two strategies, avoiding the loss of potentially more optimal allocation paths due to over-reliance on a single mapping relationship. This allows the system to maintain a dynamic balance between exploring new allocation solutions and utilizing known optimal solutions, thereby improving the robustness of the allocation strategy. In addition, a single mapping relationship is easily affected by environmental noise or random rewards, resulting in large fluctuations in the allocation ratio. Mean calculation can offset some of this random interference through statistical averaging, making allocation decisions more stable.

[0102] Optionally, the method further includes:

[0103] Step B1, determining power transmission efficiency and impedance matching; the power transmission efficiency is determined by the output power of the energy harvester and the load power of at least one load; the impedance matching is determined by the output impedance of the energy harvester and the load impedance of at least one load;

[0104] Step B2: Adjust the detection threshold corresponding to the target sensor according to the size relationship between the power transmission efficiency and the preset efficiency threshold, or the size relationship between the impedance matching degree and the preset matching threshold; the target sensor is used to provide parameters for determining the power transmission efficiency and the impedance matching degree.

[0105] In the embodiments of this application, the power transmission efficiency reflects the effective transfer ratio of energy from the energy harvester to the load. Lower power transmission efficiency indicates greater energy loss. The closer the impedance matching is to 1, the closer the impedance of the load and the energy harvester are to a conjugate match, resulting in less energy reflection and higher energy transmission efficiency.

[0106] When the power transmission efficiency is less than the preset efficiency threshold, it indicates that the energy transmission efficiency is low, which may be caused by sensor false alarm, load abnormality or impedance mismatch. If the efficiency drop is caused by sensor data fluctuation, the detection threshold of the sensor can be lowered, and then the sensor in the abnormal state can be given priority to obtain power distribution so that it can resume normal operation; if the efficiency drop is caused by load overload (such as multiple sensors running at high power consumption at the same time), the detection threshold of non-target sensors can be increased to give priority to powering the target sensor. Similarly, when adjusting the threshold based on matching degree, if the impedance matching degree is less than the preset matching threshold, it indicates that the impedance mismatch is serious and the energy reflection loss is large. In this case, the detection threshold also needs to be adjusted so that the target sensor can obtain more power distribution. This will not be repeated here.

[0107] By implementing the embodiments of the present application, by determining the power transmission efficiency and impedance matching, and adjusting the detection threshold corresponding to the target sensor according to the size relationship between the power transmission efficiency and the preset efficiency threshold, or the size relationship between the impedance matching and the preset matching threshold, the power distribution ratio of the sensor can be dynamically adjusted. When there are problems with the power transmission efficiency and impedance matching, the target sensor can obtain more power distribution, thereby effectively ensuring the operation of the target sensor and better monitoring the energy harvester and the load.

[0108] Figure 3 is a flow chart of an improved energy recovery and distribution process according to an embodiment of the present application;

[0109] Step 301, start; Step 302, system startup and initialization; Step 303, energy collection and transmission; Step 304, frequency-based adaptive regulation; Step 305, energy storage and management; Step 306, sensor working mode and data collection; Step 307, determine whether the generator is stopped, if so, go to step 308; Step 308, system standby; Step 309, end.

[0110] The above process primarily includes starting the generator and simultaneously activating the waste vibration energy recovery and self-powered monitoring system. Parameter initialization: Parameters are set for the piezoelectric-electromagnetic hybrid broadband energy harvester, initializing the impedance matching and energy optimization circuit parameters, setting parameters related to the intelligent energy priority allocation algorithm, and configuring the operating parameters of each sensor. Energy harvesting in the hybrid energy harvester: The piezoelectric module collects charge, and the electromagnetic module collects induced current. Energy transmission to the circuit: The electrical signals generated by the energy harvester (charge signals from the piezoelectric module and induced current signals from the electromagnetic module) are transmitted to the impedance matching and energy optimization circuit. Data acquisition and preprocessing: The FPGA's built-in high-speed ADC module collects the electrical signals and pre-processes them into digital signals. LSTM neural network prediction: Feature extraction and memory learning are performed on historical frequency data, outputting short-term frequency change trends. FPGA decision-making and control: By comparing predicted and actual values, the frequency difference is calculated and a control model is generated to control the MOSFET switching. Energy storage status monitoring: The energy storage unit management module monitors the hierarchically managed hybrid energy storage system in real time, collects real-time parameters of each energy storage component, and calculates the state of charge (SOC) and state of operation (SOH). Intelligent charge and discharge control: Preset intelligent charge and discharge strategies are implemented. Health Assessment and Maintenance: The energy storage unit management module regularly assesses the health of each component in the hybrid energy storage system. Data and Status Monitoring: Continuously collects data output by each sensor and monitors the energy status of the energy storage system. Data Importance and Energy Assessment: Analyzes the importance of collected data and dynamically adjusts its importance level based on pre-set rules. The generator's operating status is continuously monitored to determine if it has stopped operating. If the generator stops operating, the system enters standby mode.

[0111] Figure 4 is a schematic diagram of an energy recovery and distribution architecture provided in an embodiment of the present application; Figure 4The system includes a sensor failure mode control and data acquisition module 401, a data transmission and system monitoring module 402, an energy management system 403, an energy harvester module 404, and an adaptive adjustment module 405. In the sensor failure mode control and data acquisition module 401, the sensor collects analog signals and sends them to the embedded control system. The embedded control system generates control signals and sends them to the sensor. In the data transmission and system monitoring module 402, the embedded control system encodes and modulates the data and sends it to the host computer via a communication line. The host computer decodes the signals, performs data display analysis, and generates control instructions for the embedded control system. In the energy management system 403, the energy storage unit management module collects analog signals, converts them into digital signals, calculates parameters, and generates digital control signals based on the calculated and analyzed results. The digital control signals are sent to the control circuit, which converts the control signals into analog signals and sends them to the energy storage component. In the energy harvester module 404, a piezoelectric module outputs charge and an electromagnetic module outputs induced current. In the adaptive adjustment module 405, the FPGA and ADC obtain the electrical signal of the energy harvester module 404, convert the analog signal into a digital signal, perform digital signal processing, and then send a control signal to the drive circuit. The drive circuit controls the LC matching network to modulate the impedance, thereby storing the electrical energy in the energy storage component or distributing it to the load.

[0112] In summary, this application has the following beneficial effects:

[0113] 1. Achieve broadband energy capture. Compared with traditional single-band energy capture technology, this solution significantly improves energy capture efficiency, effectively utilizes the generator's full-band vibration energy, reduces energy waste, reduces dependence on external power supply, and meets the energy needs of the self-powered monitoring system.

[0114] 2. Intelligent Impedance Matching Optimization: In the frequency adaptive adjustment module, a dynamic control strategy based on FPGA and LSTM enables millisecond-level adaptive adjustment of LC matching network parameters. By real-time monitoring of vibration frequency and predicting trends, the matching efficiency between the system input impedance and the energy harvester output impedance is improved. Compared with traditional fixed parameter matching methods, energy transmission loss is reduced, further improving energy recovery efficiency.

[0115] 3. Accurate data monitoring to ensure system operation reliability

[0116] Dual-Q Learning Energy Allocation: A dynamic priority model based on dual-Q learning reduces the overestimation of action values ​​in traditional Q-learning algorithms and improves the accuracy of energy allocation decisions. In multi-sensor collaborative scenarios, energy can be precisely allocated based on real-time demand, avoiding missing monitoring data or system failures caused by improper energy allocation, thereby improving the reliability of the entire monitoring system.

[0117] 4. Intelligent energy management to extend the life of energy storage system

[0118] Hierarchical energy storage and health maintenance: The hierarchically managed hybrid energy storage system (graphene composite supercapacitor + lithium-ion capacitor-micro flywheel) is combined with a real-time health status assessment mechanism to precisely control the charging and discharging of energy storage components through multi-parameter monitoring such as voltage, current, and temperature.

[0119] Transient buffering optimization: The lithium-ion capacitor-micro flywheel energy storage combination responds quickly to transient high power demands, effectively protecting the main energy storage unit, reducing damage to energy storage components caused by high current shocks, and further improving the stability and service life of the energy storage system.

[0120] 5. Reduce operating costs and improve economic benefits

[0121] Self-powered design reduces wiring and maintenance: The system is fully self-powered, eliminating the need for external power lines and reducing installation costs. It also reduces maintenance work caused by external power supply failures, resulting in an estimated reduction in annual maintenance costs.

[0122] Efficient energy utilization reduces energy costs: Through efficient energy recovery and precise energy management, the overall energy consumption of the system is significantly reduced. Compared with traditional power supply methods under the same monitoring task, it has significant economic benefits.

[0123] 6. Modular design to enhance system compatibility and scalability

[0124] Hardware modular structure: Each functional module of the system (energy harvester, frequency regulation, energy management, etc.) adopts a standardized interface design and can be flexibly combined and replaced according to different generator models and monitoring requirements.

[0125] Software algorithm upgradeability: The software system based on FPGA and algorithms such as LSTM and Dual-Q Learning supports online upgrades and optimization. As monitoring requirements change and algorithms improve, system performance can be improved through remote program updates without replacing hardware equipment, reducing system upgrade costs.

[0126] Figure 5 An energy recovery and distribution device provided in an embodiment of the present application includes:

[0127] The signal conversion module 501 is used to obtain the electrical signal of the energy harvester when the energy harvester is used to collect the vibration energy of the vehicle to generate electrical energy, and convert the electrical signal into a digital signal;

[0128] a frequency prediction module 502 for inputting the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model and determine a current vibration frequency of the vehicle;

[0129] an impedance correction module 503, configured to adjust the capacitance of an impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency; the impedance matching network is connected to the energy harvester;

[0130] The electric energy distribution module 504 is configured to distribute the electric energy generated by the energy harvester to at least one load.

[0131] Optionally, the device further comprises:

[0132] a parameter calculation module, configured to determine output power, load power, output impedance, and load impedance based on voltages and currents corresponding to the energy harvester and each load, respectively, and to determine a current power transmission efficiency based on the output power and the input power;

[0133] a parameter comparison module, configured to compare the current power transmission efficiency with a historical power transmission efficiency, and to compare the output impedance with the load impedance;

[0134] An updating training module is used to input the current power transmission efficiency, the output impedance and the load impedance into the pre-training model for training when the current power transmission efficiency is less than the historical power transmission efficiency or the output impedance and the load impedance do not match.

[0135] Optional power distribution module, including:

[0136] a first distribution submodule, configured to distribute the electric energy generated by the energy harvester to at least one load and store the electric energy generated by the energy harvester in an energy storage component when the electric energy generated by the energy harvester is greater than the electric energy consumption of at least one load;

[0137] The second distribution submodule is configured to distribute the electric energy generated by the energy harvester and the electric energy in the energy storage assembly to at least one load when the electric energy generated by the energy harvester is less than the electric energy consumption of at least one load.

[0138] Optionally, the energy storage component includes a first energy storage component and a second energy storage component; the power distribution module includes:

[0139] a first charging submodule, configured to store the electrical energy generated by the energy harvester in the first energy storage component based on a first preset charge threshold and a first preset voltage threshold of the first energy storage component until the first energy storage component is fully charged; the first preset charge threshold is configured to trigger charging of the first energy storage component with a first preset current; and the first preset voltage threshold is configured to trigger charging of the first energy storage component with a first preset voltage;

[0140] The second charging submodule is used to store the electric energy generated by the energy harvester into the second energy storage component based on the second preset charge threshold and the second preset voltage threshold of the second energy storage component when the first energy storage component is fully charged, until the second energy storage component is fully charged; the second preset charge threshold is used to trigger charging of the second energy storage component with a second preset current; the second preset voltage threshold is used to trigger charging of the second energy storage component with a second preset voltage.

[0141] Optionally, the load is a sensor, and each sensor corresponds to its own detection threshold; the power distribution module includes:

[0142] A sensor threshold submodule, configured to determine the parameter value detected by each sensor and compare the parameter value with the detection threshold corresponding to the sensor;

[0143] a state combination submodule, configured to determine that the sensor is in an abnormal state when the parameter value is greater than the detection threshold, or to determine that the sensor is in a normal state when the parameter value is less than or equal to the detection threshold, and further determine a state combination corresponding to at least one sensor;

[0144] The ratio query submodule is used to determine the distribution ratio corresponding to the state combination in the pre-built mapping relationship, and distribute the electric energy generated by the energy harvester to at least one sensor based on the distribution ratio; the distribution ratio represents the proportion of electric energy corresponding to each sensor.

[0145] Optionally, the mapping relationship includes a first mapping relationship and a second mapping relationship; the ratio query submodule includes:

[0146] a respective mapping unit, configured to determine, in the first mapping relationship, at least one allocation ratio corresponding to the state combination, and in the second mapping relationship, at least one allocation ratio corresponding to the state combination; each allocation ratio is associated with an index and a value parameter; the same allocation ratio is associated with the same index and different value parameters in different mapping relationships;

[0147] a value mean unit, configured to determine, for each index, a value parameter mean corresponding to the index, to obtain at least one value parameter mean;

[0148] The target ratio unit is used to determine the maximum value parameter mean among at least one value parameter mean, and determine the target index corresponding to the maximum value parameter mean, and the target allocation ratio corresponding to the target index.

[0149] Optionally, the device further comprises:

[0150] a target parameter module for determining power transmission efficiency and impedance matching; the power transmission efficiency is determined by the output power of the energy harvester and the load power of at least one load; the impedance matching is determined by the output impedance of the energy harvester and the load impedance of at least one load;

[0151] A target sensing module is used to adjust the detection threshold corresponding to the target sensor according to the size relationship between the power transmission efficiency and a preset efficiency threshold, or the size relationship between the impedance matching degree and a preset matching threshold; the target sensor is used to provide parameters for determining the power transmission efficiency and the impedance matching degree.

[0152] In summary, in the embodiment of the present application, when the vibration energy of the vehicle is collected by an energy harvester to generate electrical energy, the electrical signal of the energy harvester is obtained and converted into a digital signal, and the digital signal is input into a pre-trained model to obtain the predicted vibration frequency output by the pre-trained model, and the current vibration frequency of the generator is determined. According to the difference between the predicted vibration frequency and the current vibration frequency, the capacitance size of the impedance matching network is adjusted to adjust the impedance, and the electrical energy generated by the energy harvester is distributed to at least one load. When collecting the vibration energy of the vehicle, the vibration frequency can be predicted and compared with the actual current frequency, the impedance can be adjusted, and then the electrical energy can be distributed, thereby improving the efficiency of power transmission.

[0153] The present application also provides an electronic device, such as Figure 6 As shown, it includes a processor 1001 , a communication interface 1002 , a memory 1003 and a communication bus 1004 , wherein the processor 1001 , the communication interface 1002 , and the memory 1003 communicate with each other via the communication bus 1004 .

[0154] The memory 1003 is used to store computer programs.

[0155] When the processor 1001 is used to execute the program stored in the memory 1003, the steps in the above-mentioned energy recovery and distribution method are implemented, which will not be repeated here.

[0156] The communication bus mentioned in the electronic device mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0157] The communication interface is used for communication between the above electronic device and other devices.

[0158] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0159] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0160] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the energy recovery and distribution method described in the above embodiment.

[0161] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute the energy recovery and distribution method described in the above embodiment.

[0162] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0163] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0164] Each embodiment in this specification is described in a related manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences from other embodiments. The embodiments of the apparatus, electronic device, computer-readable storage medium, and computer program product containing instructions thereof are generally similar to the method embodiments, so their description is relatively simple. For related portions, reference can be made to the description of the method embodiments.

[0165] The above description is only a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the scope of protection of the present application.

Claims

1. An energy recovery and distribution method, characterized in that: The method comprises: In the case of collecting vibration energy of the vehicle through an energy harvester to generate electrical energy, obtaining an electrical signal from the energy harvester and converting the electrical signal into a digital signal; Inputting the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model and determining a current vibration frequency of the vehicle; According to the difference between the predicted vibration frequency and the current vibration frequency, adjusting the capacitance of the impedance matching network to adjust the impedance; the impedance matching network is connected to the energy harvester; The electrical energy generated by the energy harvester is distributed to at least one load.

2. The method according to claim 1, characterized in that The load is a sensor, and each sensor corresponds to its own detection threshold; the step of distributing the electric energy generated by the energy harvester to at least one load includes: Determining a parameter value detected by each sensor, and comparing the parameter value with a detection threshold corresponding to the sensor; When the parameter value is greater than the detection threshold, the sensor is determined to be in an abnormal state, or when the parameter value is less than or equal to the detection threshold, the sensor is determined to be in a normal state, and then the state combination corresponding to at least one sensor is determined; In the pre-built mapping relationship, a distribution ratio corresponding to the state combination is determined, and based on the distribution ratio, the electric energy generated by the energy harvester is distributed to at least one sensor; the distribution ratio represents the proportion of electric energy corresponding to each sensor.

3. The method according to claim 2, characterized in that The mapping relationship includes a first mapping relationship and a second mapping relationship; the step of determining the allocation ratio corresponding to the state combination in the pre-constructed mapping relationship includes: In the first mapping relationship, at least one allocation ratio corresponding to the state combination is determined, and in the second mapping relationship, at least one allocation ratio corresponding to the state combination is determined; each allocation ratio is associated with an index and a value parameter; the same allocation ratio in different mapping relationships is associated with the same index and different value parameters; For each index, determining a value parameter mean corresponding to the index to obtain at least one value parameter mean; A maximum value parameter mean among at least one value parameter mean is determined, and a target index corresponding to the maximum value parameter mean and a target allocation ratio corresponding to the target index are determined.

4. The method according to claim 2, characterized in that The method further comprises: Determining power transmission efficiency and impedance matching; the power transmission efficiency is determined by the output power of the energy harvester and the load power of at least one load; the impedance matching is determined by the output impedance of the energy harvester and the load impedance of at least one load; Adjust the detection threshold corresponding to the target sensor according to the size relationship between the power transmission efficiency and the preset efficiency threshold, or the size relationship between the impedance matching degree and the preset matching threshold; the target sensor is used to provide parameters for determining the power transmission efficiency and the impedance matching degree.

5. The method according to claim 1, wherein After the step of adjusting the capacitance of the impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency, the method further includes: Determining output power, load power, output impedance, and load impedance based on voltages and currents corresponding to the energy harvester and each load, respectively, and determining a current power transmission efficiency based on the output power and the input power; comparing the current power transmission efficiency with a historical power transmission efficiency, and comparing the output impedance with the load impedance; When the current power transmission efficiency is less than the historical power transmission efficiency, or the output impedance and the load impedance do not match, the current power transmission efficiency, the output impedance and the load impedance are input into the pre-training model for training.

6. The method according to claim 1, characterized in that The step of distributing the electric energy generated by the energy harvester to at least one load comprises: In the case where the electric energy generated by the energy harvester is greater than the electric energy consumption of at least one load, the electric energy generated by the energy harvester is distributed to at least one load, and the electric energy generated by the energy harvester is stored in an energy storage component; In a case where the electric energy generated by the energy harvester is less than the electric energy consumption of at least one load, the electric energy generated by the energy harvester and the electric energy in the energy storage assembly are distributed to the at least one load.

7. The method according to claim 6, characterized in that The energy storage assembly includes a first energy storage assembly and a second energy storage assembly; the step of storing the electric energy generated by the energy harvester in the energy storage assembly includes: Based on a first preset charge threshold and a first preset voltage threshold of the first energy storage component, the electrical energy generated by the energy harvester is stored in the first energy storage component until the first energy storage component is fully charged; the first preset charge threshold is used to trigger charging of the first energy storage component with a first preset current; the first preset voltage threshold is used to trigger charging of the first energy storage component with a first preset voltage; When the first energy storage component is fully charged, the electric energy generated by the energy harvester is stored in the second energy storage component based on the second preset charge threshold and the second preset voltage threshold of the second energy storage component until the second energy storage component is fully charged; the second preset charge threshold is used to trigger charging of the second energy storage component with a second preset current; the second preset voltage threshold is used to trigger charging of the second energy storage component with a second preset voltage.

8. An energy recovery and distribution device, characterized in that: The device comprises: A signal conversion module, configured to obtain an electrical signal from an energy harvester when the energy harvester is used to harvest vibration energy of the vehicle to generate electrical energy, and convert the electrical signal into a digital signal; a frequency prediction module, configured to input the digital signal into a pre-trained model to obtain a predicted vibration frequency output by the pre-trained model and determine a current vibration frequency of the vehicle; an impedance correction module, configured to adjust the capacitance of an impedance matching network to adjust the impedance according to the difference between the predicted vibration frequency and the current vibration frequency; the impedance matching network is connected to the energy harvester; The electric energy distribution module is used to distribute the electric energy generated by the energy harvester to at least one load.

9. An electronic device, characterized in that: include: A processor, a communication interface, a memory, and a communication bus; wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; Memory for storing computer programs; The processor is configured to implement the steps of the vehicle fault diagnosis method according to any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the vehicle fault diagnosis method according to any one of claims 1 to 7 are implemented.

Images