A control method and system for low-voltage micro energy storage

Abstract

The application discloses a low-voltage micro energy storage control method and system, relates to the technical field of energy storage control, and comprises the following steps: defining physical components of low-voltage micro energy storage as nodes, defining energy transmission paths between the nodes as edges, constructing an energy flow network, using a breadth-first search algorithm to traverse the energy flow network, generating an adjacency matrix, calculating a manifold field curvature based on a power flow adjustment factor, uniformly dividing the manifold field curvature into intervals to obtain discrete states, generating control parameters for each discrete state, using a power limiting method to calculate the estimated power of the control parameters, adjusting the control parameters using the power limiting method, and generating final current control parameters. Through multi-source data acquisition and energy flow network construction, the application improves energy transmission efficiency and system stability, and through manifold field control of heat diffusion and impedance cooperation, the energy transmission efficiency of the low-voltage micro energy storage system is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy storage control, in particular to a control method and system for low-voltage micro energy storage. BACKGROUND

[0002] With the continuous evolution of distributed energy technology and the widespread deployment of micro-grid architecture, low-voltage micro energy storage systems, as key equipment to support local autonomy, peak load shifting and energy resilience, are widely used in intelligent buildings, edge computing nodes and Internet of Things energy supply platforms. Traditional low-voltage micro energy storage control methods are mostly based on static topology and preset power scheduling strategies, relying on fixed parameter PID control, linear mapping relationship between voltage, current and state of charge, and are difficult to dynamically reflect the operating conditions under local thermal disturbance, electromagnetic wave interference and complex energy interaction conditions.

[0003] The prior art still has significant deficiencies in the control of low-voltage micro energy storage systems. Traditional control methods usually rely on only a single or small number of parameters, lack comprehensive analysis of multi-source data, and the dynamic characteristics of energy flow and heat diffusion cannot be fully captured, especially when the load suddenly changes or the ambient temperature fluctuates, the control accuracy decreases. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a control method and system for low-voltage micro energy storage, which solves the problem that traditional control methods usually rely on only a single or small number of parameters, lack comprehensive analysis of multi-source data, and the dynamic characteristics of energy flow and heat diffusion cannot be fully captured, especially when the load suddenly changes or the ambient temperature fluctuates, the control accuracy decreases.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a control method for low-voltage micro energy storage, comprising,

[0008] Collecting multi-source data and preprocessing, defining the physical components of low-voltage micro energy storage as nodes and the energy transmission paths between nodes as edges, constructing an energy flow network, using a breadth-first search algorithm to traverse the energy flow network, generating an adjacency matrix, calculating the energy transmission efficiency of the energy flow network, using a gradient descent method to adjust the interaction intensity, recalculating the edge weights based on the adjusted interaction intensity, and generating a final adjacency matrix according to the new edge weights;

[0009] Calculate the local thermal diffusion rate, set the reference thermal diffusion rate using experimental calibration methods, calculate the final power upper limit, calculate the smooth current change rate using moving average, calculate the power flow adjustment factor based on the final adjacency matrix, and calculate the manifold field curvature based on the power flow adjustment factor.

[0010] The curvature of the manifold field is uniformly divided into discrete states. Control parameters are generated for each discrete state. The power of the control parameters is estimated using the power limiting method. The power limiting method is then used to adjust the parameters to generate the final current control parameters. Finally, the control parameters are converted into PWM signals using the pulse width modulation method and then transmitted and executed.

[0011] As a preferred embodiment of the control method for low-pressure micro-energy storage described in this invention, the step of calculating the energy transfer efficiency of the energy flow network and generating the final adjacency matrix includes:

[0012] Discrete wavelet transform is used to extract the low-frequency components of the electromagnetic signal, the energy spectral density of the low-frequency components is calculated, the thermal convection velocity is calculated based on three-dimensional temperature, the average thermal convection velocity is calculated, and the interaction intensity is calculated based on the energy spectral density and the average thermal convection velocity.

[0013] The physical components of low-pressure micro energy storage are defined as nodes, and the energy transfer paths between nodes are defined as edges. Based on the interaction strength, the edge weights are calculated to construct an energy flow network. The energy flow network is traversed using a breadth-first search algorithm to generate an adjacency matrix.

[0014] Calculate the energy transfer efficiency of the energy flow network, set a monitoring threshold using a statistical threshold, compare the energy transfer efficiency with the monitoring threshold, and trigger an adjustment if the energy transfer efficiency is less than the monitoring threshold. Use the gradient descent method to adjust the interaction intensity.

[0015] The edge weights are recalculated based on the adjusted interaction strength, and the final adjacency matrix is ​​generated according to the new edge weights.

[0016] As a preferred embodiment of the control method for low-pressure micro energy storage described in this invention, the step of setting a reference thermal diffusion rate using an experimental calibration method and calculating the final power upper limit includes:

[0017] Based on two-dimensional temperature, calculate the local thermal diffusion rate and the average thermal diffusion rate.

[0018] The reference thermal diffusion rate is set using an experimental calibration method. The average thermal diffusion rate is compared with the reference thermal diffusion rate. If the average thermal diffusion rate is greater than the reference thermal diffusion rate, the power upper limit of the thermal diffusion rate and ambient temperature at the next moment is calculated using a linear adjustment method. The minimum value is selected as the final power upper limit using the minimum value selection method. Otherwise, the operation continues.

[0019] As a preferred embodiment of the control method for low-pressure micro-energy storage described in this invention, the step of calculating the manifold field curvature based on the final adjacency matrix includes:

[0020] The equivalent resistance is calculated using Ohm's law, the smoothing current rate of change is calculated using moving average, and the power flow adjustment factor is calculated based on the final adjacency matrix.

[0021] The curvature of the manifold field is calculated based on the power flow adjustment factor.

[0022] As a preferred embodiment of the control method for low-voltage micro energy storage described in this invention, the step of calculating the estimated power of the control parameters using the power limiting method to generate the final current control parameters includes:

[0023] The curvature of the manifold field is uniformly divided into discrete states. Control parameters are generated for each discrete state, and the power of the control parameters is estimated using the power limiting method.

[0024] The estimated power is compared with the final power limit. If the estimated power is greater than the final power limit, the power limiting method is used for adjustment to generate the final current control parameters; otherwise, operation continues.

[0025] As a preferred embodiment of the control method for low-voltage micro energy storage described in this invention, the step of converting control parameters into PWM signals and transmitting and executing them using a pulse width modulation method includes:

[0026] The final current control parameters are normalized, and the switching frequency control parameters are mapped to the PWM frequency using the direct mapping method. The normalized final current control parameters and PWM frequency are converted into PWM signals using the pulse width modulation method. The PWM signals are then transmitted and executed using the I2C protocol.

[0027] As a preferred embodiment of the control method for low-voltage micro-energy storage described in this invention, the step of collecting and preprocessing multi-source data includes:

[0028] Based on smart sensors, downsampling is used to collect multi-source data of low-voltage micro energy storage at a uniform frequency, and data time synchronization, noise reduction and normalization are performed.

[0029] The intelligent sensors include infrared thermal imaging, fiber optic temperature, thermocouples, voltage, current, and Hall effect sensors.

[0030] The multi-source data includes two-dimensional temperature, three-dimensional temperature, ambient temperature, voltage, current, and electromagnetic signal data.

[0031] Secondly, the present invention provides a control system for low-voltage micro energy storage, comprising,

[0032] The network collection module is used to collect multi-source data and perform preprocessing. The physical components of low-voltage micro-energy storage are defined as nodes, and the energy transfer paths between nodes are defined as edges. An energy flow network is constructed, and a breadth-first search algorithm is used to traverse the energy flow network to generate an adjacency matrix. The energy transfer efficiency of the energy flow network is calculated, and the interaction strength is adjusted using the gradient descent method. The edge weights are recalculated based on the adjusted interaction strength, and the final adjacency matrix is ​​generated based on the new edge weights.

[0033] The power curvature module is used to calculate the local thermal diffusion rate, set the reference thermal diffusion rate using an experimental calibration method, calculate the final power upper limit, calculate the smooth current change rate using a moving average, calculate the power flow adjustment factor based on the final adjacency matrix, and calculate the manifold field curvature based on the power flow adjustment factor.

[0034] The control execution module is used to divide the curvature of the manifold field into uniform intervals to obtain discrete states, generate control parameters for each discrete state, calculate the estimated power of the control parameters using the power limiting method, adjust them using the power limiting method, generate the final current control parameters, and convert the control parameters into PWM signals using the pulse width modulation method for transmission and execution.

[0035] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the control method for low-voltage micro-energy storage as described in the first aspect of the present invention.

[0036] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the control method for low-voltage micro-energy storage as described in the first aspect of the present invention.

[0037] The beneficial effects of this invention are as follows: This invention improves energy transfer efficiency and system stability through multi-source data acquisition and energy flow network construction, and enhances the energy transmission efficiency of low-pressure micro energy storage systems through manifold field control that combines thermal diffusion and impedance. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of the control method for low-voltage micro energy storage in Example 1.

[0040] Figure 2 This is a schematic diagram of the control system for low-voltage micro energy storage in Example 1. Detailed Implementation

[0041] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0043] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0044] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a control method for low-voltage micro energy storage, including the following steps:

[0045] S1. Collect and preprocess multi-source data, define the physical components of low-pressure micro energy storage as nodes, define the energy transfer paths between nodes as edges, construct an energy flow network, traverse the energy flow network using a breadth-first search algorithm, generate an adjacency matrix, calculate the energy transfer efficiency of the energy flow network, adjust the interaction strength using gradient descent, recalculate the edge weights based on the adjusted interaction strengths, and generate the final adjacency matrix based on the new edge weights.

[0046] Specifically, this involves collecting and preprocessing multi-source data, including:

[0047] Based on smart sensors, downsampling is used to collect multi-source data of low-voltage micro energy storage at a uniform frequency, and data time synchronization, noise reduction and normalization are performed.

[0048] The intelligent sensors include infrared thermal imaging, fiber optic temperature, thermocouples, voltage, current, and Hall effect sensors.

[0049] The multi-source data includes two-dimensional temperature, three-dimensional temperature, ambient temperature, voltage, current, and electromagnetic signal data.

[0050] By using intelligent sensor networks to comprehensively and multidimensionally sense energy storage units, data collection covering physical quantities such as heat, electricity, and magnetism is carried out. The downsampling method to unify the sampling frequency not only avoids information redundancy caused by data dimensional imbalance, but also reduces system load, ensures the spatiotemporal consistency of data, and effectively avoids time-series drift caused by inconsistent sampling frequencies between sensors.

[0051] Furthermore, the energy transfer efficiency of the energy flow network is calculated to generate the final adjacency matrix, including:

[0052] The low-frequency components of the electromagnetic signal are extracted using discrete wavelet transform, and the energy spectral density of the low-frequency components is calculated using the following formula:

[0053] ;

[0054] in Where N is the energy density of the electromagnetic signal, and N is the number of sampling points. Let r be the r-th approximation coefficient, representing the low-frequency component;

[0055] Based on three-dimensional temperature, the thermal convection velocity is calculated using the following formula:

[0056] ;

[0057] ;

[0058] in Let t be the thermal convection velocity at the spatial point (x, y, z). , as well as These are the electrolyte density, specific heat capacity, and thermal conductivity, respectively, provided by the equipment manufacturer. The three-dimensional temperature-time derivative is calculated using the backward difference method. Let t be the three-dimensional temperature at a spatial point (x, y, z). The sampling interval is set using a fixed time interval method;

[0059] The formula for calculating the average thermal convection velocity is:

[0060] ;

[0061] in The average thermal convection velocity over time t is X, Y, and Z are the three-dimensional temperature field grid sizes, set using a uniform grid method.

[0062] The interaction intensity is calculated based on the energy spectral density and the average thermal convection velocity, using the following formula:

[0063] ;

[0064] in For interaction strength, and These are the reference electromagnetic signal energy density and the reference thermal convection velocity, respectively, set using experimental calibration methods.

[0065] The physical components of low-voltage micro-energy storage are defined as nodes, including battery cells and load cells. The energy transfer paths between nodes are defined as edges, based on physical connections including electrical connections (such as wires from the battery to the processor) and thermal conduction paths (such as heat flow from the battery to the heat sink). Based on the interaction strength, edge weights are calculated to construct an energy flow network, as shown in the formula:

[0066] ;

[0067] in Let t be the edge weight between nodes i and j, where negative values ​​represent energy dissipation and positive values ​​represent energy transfer. and The power of nodes i and j is calculated based on the product of voltage and current.

[0068] The energy flow network is traversed using a breadth-first search algorithm to generate an adjacency matrix;

[0069] The formula for calculating the energy transfer efficiency of an energy flow network is:

[0070] ;

[0071] in Let be the energy transfer efficiency over time t. and The number of nodes consuming and supplying electrical energy is determined by the number of nodes; the current of battery nodes is greater than 0, and the current of load nodes is less than 0.

[0072] A statistical threshold is used to set the monitoring threshold. The energy transfer efficiency is compared with the monitoring threshold. If the energy transfer efficiency is less than the monitoring threshold, an adjustment is triggered. The interaction intensity is adjusted using the gradient descent method, with the following formula:

[0073] ;

[0074] in Let t be the time, and t be the corrected interaction strength. The learning rate is set using a web search method. To achieve the target efficiency, a fixed calibration method is used for setting. The partial derivative of energy transfer efficiency with respect to interaction strength is calculated using numerical difference.

[0075] The edge weights are recalculated based on the adjusted interaction strength. The final adjacency matrix is ​​then generated using the new edge weights, as shown in the formula:

[0076] ;

[0077] in Let be the final adjacency matrix at time t. Let the weight of the new edge be at time t. These are the elements of the initial adjacency matrix, where 1 indicates that an edge exists and 0 indicates that it does not exist.

[0078] Temperature, electromagnetic, and flow field signals have different response delays and time-domain distributions. Fusion processing can bridge the control lag problem caused by physical information asymmetry. Regularized data structures help form sparse graph structures, improving the processing efficiency of adjacency matrices and reducing computational complexity. Low-frequency components of electromagnetic disturbances better reflect the steady-state power disturbances of the system; separating them from high-frequency interference enhances the physical meaning of the control signal. Daubechies wavelets outperform Fourier transforms in handling signal abrupt changes and boundary preservation, making them suitable for time-varying analysis of perturbation systems. Thermal convection velocity is the natural diffusion rate of heat loss within the energy storage unit, revealing internal local fault trends. Normalizing and combining electromagnetic wave energy density and thermal convection velocity reflects the electromagnetic... The dual impact of disturbances and local heat accumulation avoids misjudgments caused by isolated analysis. The normalization standard is established through experimental calibration, giving various physical parameters a unified control weight meaning. The power term can reflect the instantaneous supply and demand matching situation, enhancing the system's ability to respond to sudden load changes. The adjacency matrix formed after breadth-first search (BFS) traversal has real-time and dynamic topology expression capabilities, unlike traditional static topology modeling. It adaptively adjusts the interaction intensity through gradient descent, avoiding manual parameter tuning and improving the system's intelligence level. Dynamic changes in the adjacency matrix can trigger the addition or deletion of edges, simulating the physical response effects of "virtual disconnection" and "priority channels". After reconstruction, the network can automatically avoid faulty edges, improving anti-interference and system robustness.

[0079] S2. Calculate the local thermal diffusion rate, set the reference thermal diffusion rate using experimental calibration methods, calculate the final power upper limit, calculate the smooth current change rate using moving average, calculate the power flow adjustment factor based on the final adjacency matrix, and calculate the manifold field curvature based on the power flow adjustment factor.

[0080] Specifically, an experimental calibration method is used to set a reference thermal diffusion rate and calculate the final power upper limit, including:

[0081] Based on two-dimensional temperature, the local thermal diffusion rate is calculated using the following formula:

[0082] ;

[0083] in The thermal diffusion rate at time t and location (x, y) is calculated using the backward difference method. Let t be the two-dimensional temperature at position (x, y);

[0084] The formula for calculating the average thermal diffusion rate is:

[0085] ;

[0086] in R is the average thermal diffusion rate at time t, m is the time window index, R and N are the temperature field grid sizes for two-dimensional temperature, set using a uniform grid division method, and M is the size of the time window, set using a sliding window method.

[0087] A reference thermal diffusion rate is set using an experimental calibration method. The average thermal diffusion rate is compared with the reference rate. If the average rate is greater than the reference rate, the power upper limit for the next moment is calculated using a linear adjustment method based on the thermal diffusion rate and ambient temperature. The minimum value is selected as the final power upper limit using a minimum value selection method. Otherwise, the operation continues. The formula is:

[0088] ;

[0089] ;

[0090] in and These represent the upper limits of power for the next moment's thermal diffusion rate and ambient temperature, respectively. For reference thermal diffusion rate, To adjust the coefficients, the heat conduction model method was used. The basic reduction factor is set using the thermal stability method. The temperature deviation coefficient is set using the linear thermal expansion method. Let t be the ambient temperature. The ambient temperature threshold was set using experimental calibration methods.

[0091] By extracting local temperature gradient changes in real time using the backward difference method, the system can predict rapidly heating regions and intervene early to adjust the power distribution path, reducing the risk of thermal stress concentration. Using a two-dimensional temperature distribution provides a more comprehensive thermal diffusion map than point monitoring, which can be used to identify design defects in the heat dissipation structure and improve hardware tuning efficiency. By calculating the time series of thermal diffusion rate through moving average, the system can fit the thermal diffusion curves of steady and unstable states, providing a mathematical basis for subsequent thermal runaway prediction. Coupled with a heat-power model, the system improves control accuracy. With the addition of thermal diffusion constraints, a cross-domain energy control strategy can be implemented, thereby fully utilizing the system's electrical performance while ensuring thermal safety. By introducing parameters such as temperature deviation coefficient and basic reduction factor, the power limit boundary is smoothed linearly, effectively reducing mechanical fatigue and electrochemical instability caused by drastic power fluctuations.

[0092] Furthermore, based on the final adjacency matrix, the manifold field curvature is calculated, including:

[0093] The equivalent resistance is calculated using Ohm's law, and the smoothed rate of change of current is calculated using the moving average method. The formula is as follows:

[0094] ;

[0095] in Let be the smoothed rate of change of current over time t. Let be the equivalent resistance at time t. Let be the current at time t;

[0096] Based on the final adjacency matrix, the power flow adjustment factor is calculated using the following formula:

[0097] ;

[0098] in The power flow adjustment factor at time t. It is a very small constant;

[0099] Based on the power flow adjustment factor, the manifold field curvature is calculated using the following formula:

[0100] ;

[0101] in Let be the curvature of the manifold field at time t. The reference current change rate was set using experimental calibration methods. The state of charge of the battery at time t is calculated using the coulomb counting method. To achieve the optimal state of battery charge, an energy efficiency optimization method is used.

[0102] By mapping the originally discrete energy nodes to continuous manifolds through adjacency matrices, the distribution, accumulation, and transfer of power can be expressed using geometric curvature, providing a new descriptive paradigm for energy optimization control. Composed of the current rate of change and equivalent resistance, it can significantly amplify the weight of high-resistance-high-variance regions in the curvature, enabling energy management to prioritize critical transmission paths and avoid bottleneck effects. By introducing the relative change deviation between the state of charge (SOC) and the rate of change of current to establish a curvature model, thermal, electrical, and state information are jointly embedded in the geometric structure to form a multi-field collaborative dynamic description framework. Through characterization parameters such as thermal diffusion rate and manifold curvature, the transformation from local point state identification to global surface domain energy flow dynamic modeling is completed, constructing a real-time feedback loop between power regulation and environmental variables, and enhancing the system's robustness and energy efficiency in dynamic environments.

[0103] S3. Divide the curvature of the manifold field into uniform intervals to obtain discrete states. Generate control parameters for each discrete state. Calculate the estimated power of the control parameters using the power limiting method. Adjust the parameters using the power limiting method to generate the final current control parameters. Convert the control parameters into PWM signals using the pulse width modulation method and transmit and execute them.

[0104] Specifically, the power estimation of the control parameters is calculated using the power limiting method to generate the final current control parameters, including:

[0105] The curvature of the manifold is uniformly divided into discrete intervals to obtain discrete states. Control parameters are generated for each discrete state, using the following formula:

[0106] ;

[0107] ;

[0108] in Let K be the current control parameter for the k-th discrete state, where k is the discrete state index and K is the number of partitions. The target power at time t is set using a fixed calibration method. The frequency is set using an efficiency optimization method. The switching frequency control parameters are for the k-th discrete state.

[0109] The estimated power for calculating control parameters using the power limiting method is given by the following formula:

[0110] ;

[0111] in For the estimated power at time t, The nominal voltage is set using an efficiency optimization method;

[0112] The estimated power is compared with the final power limit. If the estimated power is greater than the final power limit, the power limiting method is used for adjustment to generate the final current control parameters; otherwise, operation continues. The formula is:

[0113] ;

[0114] in For the final current control parameters of the k-th discrete state, This represents the final power limit for the next moment.

[0115] While continuous space modeling is accurate, it incurs high computational costs and is not conducive to real-time control. Each state corresponds to a set of preset control parameters, significantly reducing controller decision time. Discrete state partitioning allows for the reuse or expansion of control strategies across different application scenarios (e.g., different battery types, power levels), improving the portability of the control framework. In scenarios with strong disturbances or large changes, boundary value warning mechanisms in discrete intervals can trigger protection measures more quickly, preventing the system from entering extreme states such as thermal runaway or voltage drops. Once the estimated power exceeds the "final power limit" obtained from prior thermal diffusion and ambient temperature constraints, a limiting mechanism is triggered, effectively preventing heat accumulation and capacitor damage caused by overpower. Compared to hard-cut... The interruption mechanism and amplitude limiting method maintain output smoothness through slope adjustment and scaling transformation, which helps to avoid system oscillation and improve the response coordination of the drive module. This invention sets independent frequency control parameters, so that the current control parameters and the switching frequency have a coupled rather than overlapping relationship, which improves control flexibility and reduces modulation error. The control parameters are calculated by combining the target power, discrete state control factor and frequency characteristics, which ensures both power controllability and optimal switching efficiency, achieving coordination in the three-dimensional space of energy-heat-frequency. By configuring a set of control parameters for each discrete state, the software optimization results can be coupled and deployed with the hardware control circuit, reducing the real-time calculation burden and improving the response frequency of the power controller.

[0116] Furthermore, the control parameters are converted into PWM signals using pulse width modulation (PWM) methods for transmission and execution, including:

[0117] The final current control parameters are normalized, and the switching frequency control parameters are mapped to the PWM frequency using the direct mapping method. The normalized final current control parameters and PWM frequency are converted into PWM signals using the pulse width modulation method. The PWM signals are then transmitted to the power control module using the I2C protocol for execution.

[0118] The PWM method precisely controls the current output by changing the duty cycle, making it suitable for micropower control scenarios with extremely high requirements for adjustment accuracy. The I2C protocol supports multi-master and multi-slave structures, low-power communication, and anti-interference capabilities, making it suitable for deployment in embedded systems and multi-node energy networks. The introduction of normalization processing and mapping mechanisms effectively avoids the expansion of quantization errors, improves signal resolution accuracy, and ensures that the control signal remains synchronized under high-frequency switching.

[0119] This embodiment also provides a control system for low-voltage micro energy storage, including:

[0120] The network collection module is used to collect multi-source data and perform preprocessing. The physical components of low-voltage micro-energy storage are defined as nodes, and the energy transfer paths between nodes are defined as edges. An energy flow network is constructed, and a breadth-first search algorithm is used to traverse the energy flow network to generate an adjacency matrix. The energy transfer efficiency of the energy flow network is calculated, and the interaction strength is adjusted using the gradient descent method. The edge weights are recalculated based on the adjusted interaction strength, and the final adjacency matrix is ​​generated based on the new edge weights.

[0121] The power curvature module is used to calculate the local thermal diffusion rate, set the reference thermal diffusion rate using an experimental calibration method, calculate the final power upper limit, calculate the smooth current change rate using a moving average, calculate the power flow adjustment factor based on the final adjacency matrix, and calculate the manifold field curvature based on the power flow adjustment factor.

[0122] The control execution module is used to divide the curvature of the manifold field into uniform intervals to obtain discrete states, generate control parameters for each discrete state, calculate the estimated power of the control parameters using the power limiting method, adjust them using the power limiting method, generate the final current control parameters, and convert the control parameters into PWM signals using the pulse width modulation method for transmission and execution.

[0123] This embodiment also provides a computer device applicable to the control method of low-voltage micro energy storage, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the control method of low-voltage micro energy storage as proposed in the above embodiment.

[0124] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0125] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the control method for realizing low-voltage micro energy storage as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0126] In summary, this invention improves energy transfer efficiency and system stability through multi-source data acquisition and energy flow network construction, and enhances the energy transmission efficiency of low-pressure micro energy storage systems through manifold field control that combines thermal diffusion and impedance.

[0127] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A control method of low-voltage micro energy storage, characterized by: comprising, collecting multi-source data and preprocessing, defining physical components of low-voltage micro energy storage as nodes, energy transfer paths between nodes as edges, constructing an energy flow network, using breadth-first search algorithm to traverse the energy flow network, generating an adjacency matrix, calculating the energy transfer efficiency of the energy flow network, adjusting the interaction intensity using gradient descent method, recalculating the edge weight based on the adjusted interaction intensity, generating the final adjacency matrix according to the new edge weight; calculating the local heat diffusion rate, setting the reference heat diffusion rate using the experimental calibration method, calculating the final power upper limit, calculating the smoothed current change rate using the moving average method, calculating the power flow adjustment factor based on the final adjacency matrix, calculating the manifold field curvature based on the power flow adjustment factor; uniformly dividing the manifold field curvature into intervals to obtain discrete states, generating control parameters for each discrete state, calculating the estimated power of the control parameters using the power limiting method, adjusting using the power limiting method to generate the final current control parameters, converting the control parameters into PWM signals using the pulse width modulation method and transmitting and executing.

2. The control method of low voltage micro energy storage as claimed in claim 1, characterized by: The calculation of the energy transfer efficiency of the energy flow network, the generation of the final adjacency matrix, comprises: extracting the low-frequency component of the electromagnetic signal using discrete wavelet transform, calculating the energy spectrum density of the low-frequency component, calculating the heat convection velocity based on the three-dimensional temperature, calculating the average heat convection velocity, calculating the interaction intensity based on the energy spectrum density and the average heat convection velocity; defining the physical components of low-voltage micro energy storage as nodes, defining the energy transfer paths between nodes as edges, calculating the edge weight based on the interaction intensity, constructing the energy flow network, using breadth-first search algorithm to traverse the energy flow network, generating an adjacency matrix; calculating the energy transfer efficiency of the energy flow network, setting the monitoring threshold using the statistical threshold, comparing the energy transfer efficiency with the monitoring threshold, if the energy transfer efficiency is less than the monitoring threshold, triggering adjustment, adjusting the interaction intensity using the gradient descent method; recalculating the edge weight based on the adjusted interaction intensity, generating the final adjacency matrix according to the new edge weight.

3. The control method of low voltage micro energy storage according to claim 2, characterized in that: The use of experimental calibration method to set the reference heat diffusion rate and calculate the final power upper limit, comprises: calculating the local heat diffusion rate based on the two-dimensional temperature, calculating the average heat diffusion rate; using the experimental calibration method to set the reference heat diffusion rate, comparing the average heat diffusion rate with the reference heat diffusion rate, if the average heat diffusion rate is greater than the reference heat diffusion rate, calculating the power upper limit of the next time heat diffusion rate and environmental temperature using the linear adjustment method, selecting the minimum value as the final power upper limit using the minimum value selection method, otherwise continue to run.

4. The control method of low voltage micro energy storage according to claim 3, characterized in that: The calculation of the manifold field curvature based on the final adjacency matrix, comprises: calculating the equivalent resistance using Ohm's law, calculating the smoothed current change rate using the moving average method, calculating the power flow adjustment factor based on the final adjacency matrix; calculating the manifold field curvature based on the power flow adjustment factor.

5. The control method of low voltage micro energy storage as claimed in claim 4, characterized by: The use of power limiting method to calculate the estimated power of control parameters and generate the final current control parameters, comprises: The manifold field curvature is uniformly divided into intervals to obtain a discrete state, control parameters are generated for each discrete state, and the estimated power of the control parameters is calculated using a power clipping method; The estimated power is compared with the final power upper limit. If the estimated power is greater than the final power upper limit, the power clipping method is used for adjustment to generate the final current control parameter, otherwise the operation continues.

6. The control method of a low voltage micro energy storage according to claim 5, characterized in that: The control parameters are converted into PWM signals using a pulse width modulation method and transmitted and executed, including: The final current control parameters are normalized, the switching frequency control parameters are mapped into PWM frequencies using a direct mapping method, the normalized final current control parameters and the PWM frequencies are converted into PWM signals using a pulse width modulation method, the PWM signals are transmitted using an I2C protocol and executed.

7. The control method of low voltage micro energy storage as claimed in claim 1, wherein: The multi-source data is collected and preprocessed, including: Based on intelligent sensors, the multi-source data of low-voltage micro energy storage is collected at a uniform frequency using a downsampling method, and data time synchronization, denoising and normalization processing are performed; The intelligent sensors include infrared thermal imaging, optical fiber temperature, thermocouple, voltage, current and Hall effect sensor; The multi-source data includes two-dimensional temperature, three-dimensional temperature, ambient temperature, voltage, current and electromagnetic signal data.

8. A control system of low-voltage micro energy storage, based on the control method of low-voltage micro energy storage according to any one of claims 1-7, characterized in that: It includes, The network module is used to collect multi-source data and perform preprocessing, the physical components of low-voltage micro energy storage are defined as nodes, the energy transfer path between nodes is defined as an edge, an energy flow network is constructed, a breadth-first search algorithm is used to traverse the energy flow network, an adjacency matrix is generated, the energy transfer efficiency of the energy flow network is calculated, the interaction intensity is adjusted using a gradient descent method, the edge weight is recalculated based on the adjusted interaction intensity, and the final adjacency matrix is generated according to the new edge weight; The power curvature module is used to calculate the local heat diffusion rate, set the reference heat diffusion rate using an experimental calibration method, calculate the final power upper limit, calculate the smooth current change rate using a moving average, calculate the power flow adjustment factor based on the final adjacency matrix, and calculate the manifold field curvature based on the power flow adjustment factor; The control execution module is used to uniformly divide the manifold field curvature into intervals to obtain a discrete state, generate control parameters for each discrete state, calculate the estimated power of the control parameters using a power clipping method, adjust using a power clipping method to generate the final current control parameter, and convert the control parameters into PWM signals using a pulse width modulation method and transmit and execute. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the low-voltage micro energy storage control method of any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the low-voltage micro energy storage control method of any one of claims 1-7.

Images