Method for online monitoring green energy supplement

Abstract

This invention relates to the field of new energy power supply and IoT monitoring technology, and particularly to an online monitoring method for green energy replenishment. This invention achieves deep integration and efficient utilization of multiple energy sources. By equipping each energy source with an independent MPPT primary optimizer and defining energy characteristic tags, the energy is first stabilized and characterized at the source. Then, through the dynamic priority evaluation algorithm of the intelligent energy replenishment management center, the dynamic priority value of each energy source is calculated in real time, and the primary and secondary energy sources and operating modes are intelligently arbitrated accordingly. Through the reconfigurable network and intelligent merging (such as time / phase interleaving) of the multi-input collaborative charging management module, the orderly management and efficient merging of energy sources with different characteristics are achieved. The method avoids direct conflicts and efficiency backflow between different energy sources at the source, pipe, and sink levels, resulting in a significant improvement in the overall energy capture and utilization efficiency of the system compared to traditional hybrid schemes.

Description

Technical Field

[0001] This invention relates to the field of new energy power supply and Internet of Things monitoring technology, and in particular to an online monitoring method for green energy replenishment. Background Technology

[0002] Currently, with the deep application of IoT and big data technologies in environmental monitoring, industrial operation and maintenance, smart cities, and other fields, a large number of online monitoring devices (such as water quality monitors, weather stations, pipeline sensors, and mobile monitoring terminals) are deployed in remote outdoor areas, mobile platforms, or special scenarios such as pipelines and underground. These scenarios typically lack stable and convenient mains power access, while the devices themselves need to operate continuously to complete real-time data acquisition and transmission, resulting in high power consumption. This makes long-term and stable power supply a core technical bottleneck restricting their widespread application.

[0003] Currently, the main solutions available using existing technologies are as follows:

[0004] If a pure solar power system is used, this solution is severely limited by sunlight conditions and cannot work during continuous rainy weather, at night, or in areas with insufficient sunlight, resulting in poor power supply reliability. Another solution is to use large-capacity disposable batteries or battery packs. Although deployment is flexible, the battery life is limited, requiring frequent manual replacement or charging. In remote areas, maintenance costs are extremely high, and discarded batteries create environmental pressure.

[0005] Some improvement schemes attempt to connect multiple environmental energy harvesting devices, such as solar panels, micro hydroelectric generators, and vibration generators, to the power system of the monitoring equipment simultaneously. However, this approach is essentially just a simple parallel connection. Because the output characteristics (voltage, current, fluctuations, internal resistance) of different environmental energy sources (such as light, water, and vibration) vary greatly and are extremely unstable, direct parallel connection can lead to the output of efficient energy sources being dragged down by inefficient or poorly functioning energy sources, resulting in a severe energy cannibalization effect. The overall energy capture and utilization efficiency of the system is far below the theoretical value, and voltage conflicts may even damage the power generation unit or charging circuit.

[0006] Some existing technologies incorporate simple charging management circuits, such as switching energy sources based on voltage levels. However, this control strategy is static or semi-static and cannot coordinate and optimize scheduling based on the real-time power generation capacity of each energy source, the health status of the energy storage battery and the current charging stage (e.g., constant current, constant voltage, trickle charging), and the dynamic power consumption of the load. The result is rigid energy management, unable to achieve optimal energy allocation in complex and changing environments. Batteries may remain in a suboptimal charging state for extended periods, shortening their lifespan, and the system cannot cope with sudden load changes or energy shortages.

[0007] In summary, the core technical problem of existing technologies lies in how to deeply integrate and dynamically optimize the scheduling of unstable and heterogeneous environmental energy sources such as solar energy, hydroelectric energy, and vibration energy through an intelligent method to overcome the efficiency loss problem of simple parallel connection and achieve global optimal matching with battery charging curves and load power demand. This would provide a high-efficiency, stable, adaptive, and long-life green energy replenishment solution for high-power online monitoring devices in scenarios without mains power.

[0008] This invention addresses the technical problems of low energy utilization efficiency, poor system adaptability, easily damaged battery life, and low overall power supply reliability in the existing technologies mentioned above, and proposes an online monitoring method for green energy replenishment. Summary of the Invention

[0009] The technical solution adopted by this invention to solve the above-mentioned technical problems is: an online monitoring method for green energy replenishment. This method is implemented based on an energy replenishment system, which includes a heterogeneous energy harvesting module cluster, an intelligent energy replenishment management center, a multi-input collaborative charging management module, an energy storage battery, and an online monitoring instrument load. The method specifically includes the following steps:

[0010] S1. Construct a cluster of heterogeneous energy harvesting modules that can be independently optimized. For three types of energy sources, namely solar energy, hydroelectric energy, and vibration energy, energy harvesting and preliminary stabilization output are carried out through independent power generation units and their integrated maximum power point tracking primary optimizers. A unique energy characteristic label is defined for each type of energy.

[0011] S2. Establish a multimodal collaborative perception and decision-making network. The intelligent energy replenishment management center performs real-time data acquisition, dynamic priority evaluation, arbitration and strategy generation to determine the main energy source, auxiliary energy source and their working mode in the current cycle.

[0012] S3. Execute charging management and energy distribution based on multi-input collaboration. The multi-input collaborative charging management module performs reconfigurable path management and intelligent convergence of electrical energy from the main energy source and auxiliary energy source according to the strategy instructions generated by the intelligent energy replenishment management center, so as to charge the energy storage battery.

[0013] S4. Implement adaptive closed-loop management of battery and load, including adaptive load negotiation based on energy prediction and battery health-oriented charging based on battery health status.

[0014] Furthermore, in step S1, the energy characteristic label includes at least: energy type, rated output voltage range, current instantaneous output power, power fluctuation coefficient, and expected sustainable power generation time based on historical data.

[0015] Furthermore, the dynamic priority assessment in step S2 specifically involves the intelligent energy replenishment management center calculating a dynamic priority value for each available energy source. The calculation of this dynamic priority value comprehensively considers the instantaneous output power level, power stability, expected sustainable power generation time, and the degree of matching with the current charging stage requirements of the energy storage battery.

[0016] Furthermore, the arbitration and strategy generation in step S2 also includes: when it is determined that the online monitoring instrument load is in an instantaneous high power consumption mode, and the dynamic priority value and output power of the main energy source meet the preset conditions, a strategy including the load direct compensation channel mode is generated, and the instruction is given to divert part of the electrical energy generated by the energy source to directly supply the online monitoring instrument load.

[0017] Furthermore, the reconfigurable pathway management of electrical energy from the auxiliary power source described in step S3 includes: if the output characteristics of the auxiliary power source do not match the current battery charging demand, then guiding it to a shared buffer energy storage unit for temporary energy storage and shaping.

[0018] Furthermore, the intelligent confluence in step S3 specifically refers to the following: the power path for future autonomous energy replenishment and the power released from the shared buffer energy storage unit are interleaved in time or phase to form a charging current waveform for the energy storage battery.

[0019] Furthermore, the adaptive load negotiation based on energy prediction in step S4 specifically includes: the intelligent energy replenishment management center predicts, based on the expected sustainable power generation time of each energy source and historical energy collection data, that the total energy available for replenishing the energy storage battery will be lower than the energy required to maintain the online monitoring instrument load at rated power consumption within the next preset evaluation period, and then sends an early warning to the online monitoring instrument load and negotiates to enter the graded energy saving mode.

[0020] Furthermore, the battery health-oriented charging based on battery health status in step S4 specifically includes: using the health status of the energy storage battery as one of the input parameters for calculating the dynamic priority value, and adopting a charging strategy that reduces the charging current threshold and avoids using energy with drastic fluctuations as the main supplementary energy for batteries with declining health status.

[0021] Furthermore, the heterogeneous energy harvesting module cluster includes a solar energy harvesting module, a water flow energy harvesting module, and a vibration energy harvesting module. The solar energy harvesting module adopts a monocrystalline silicon photovoltaic panel and its maximum power point tracking DC-DC optimizer. The water flow energy harvesting module adopts an integrated design of a micro turbine generator and a power optimizer. The vibration energy harvesting module adopts a piezoelectric electromagnetic composite energy harvester.

[0022] Furthermore, the multi-input collaborative charging management module contains a reconfigurable network composed of high-speed power switches, used to switch power paths according to strategy instructions; the intelligent power replenishment management hub is implemented by an ultra-low power microcontroller, used to run a dynamic priority evaluation algorithm.

[0023] The advantages of this invention are:

[0024] 1. This invention achieves deep integration and efficient utilization of multiple energy sources. By equipping each energy source with an independent MPPT primary optimizer and defining energy characteristic tags, the energy is initially stabilized and characterized at the source. Then, through the dynamic priority evaluation algorithm of the intelligent energy replenishment management center, the dynamic priority value of each energy source is calculated in real time, and the primary and secondary replenishment energy sources and operating modes are intelligently arbitrated accordingly. Through the reconfigurable network and intelligent merging (such as time / phase interleaving) of the multi-input collaborative charging management module, the orderly management and efficient merging of energy sources with different characteristics are achieved. This method avoids direct conflicts and efficiency backflow between different energy sources at the source, pipeline, and sink levels, resulting in a significant improvement in the overall energy capture and utilization efficiency of the system compared to traditional hybrid schemes.

[0025] 2. The method of this invention does not rely on fixed energy priorities. Its dynamic priority evaluation algorithm comprehensively considers instantaneous power, power stability, expected sustainable power generation time, and the degree of matching with the current charging needs of the battery, enabling the system to adaptively select the most suitable energy replenishment strategy based on real-time environmental energy conditions (such as changes in light intensity, fluctuations in water flow rate, and the presence or absence of vibration) and the system's own state (battery SOC, load power consumption). In particular, the direct load replenishment channel mode can directly utilize high-quality energy to divert power supply when the load suddenly increases, effectively coping with the challenge of instantaneous high power consumption and enhancing the system's continuous working capability in complex and variable environments.

[0026] 3. This invention introduces a charging strategy adjustment based on battery health, which can implement protective charging measures for aging batteries to avoid overcharging, over-discharging, and inappropriate charging current. Simultaneously, through adaptive load negotiation based on energy prediction, the system can proactively negotiate with load devices to enter a tiered energy-saving mode when energy shortages are anticipated. These two mechanisms finely manage energy flow from both charging and usage dimensions, maximizing the cycle life of high-performance energy storage batteries, reducing maintenance frequency due to battery failure, and lowering overall lifecycle operating costs.

[0027] 4. This invention comprehensively covers the intelligent hybrid utilization of solar energy, water flow energy, and vibration energy, and can be flexibly adapted to various working conditions such as outdoor sunlight, river pipes, and vibrating equipment. Its core intelligent management logic can be implemented through software algorithms, eliminating the need for laying long-distance power lines and making installation and deployment convenient. This method fundamentally solves the problem of long-term power supply for online monitoring equipment in outdoor, mobile, and off-grid scenarios, providing a reliable energy guarantee for the promotion of various IoT monitoring applications. Detailed Implementation

[0028] The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Example 1:

[0030] This embodiment uses a river water quality online monitoring station deployed in an area without mains power as an application scenario to illustrate the implementation process of the method of the present invention in detail. The monitoring station's load includes a multi-parameter water quality sensor (monitoring pH, dissolved oxygen, turbidity, conductivity, etc.), a data acquisition processor, a 4G / 5G wireless communication module, and auxiliary components. The system's average power consumption is approximately 5W, and the instantaneous peak power consumption during wireless data transmission can reach 15W. The site has various environmental energy sources, including solar energy, water flow energy (river water flow), and vibration energy (water flow impacting the equipment foundation and wind-induced vibration).

[0031] In this embodiment, the cluster comprises three independent power generation units with built-in primary optimizations. The energy replenishment system is integrated and installed within the monitoring station. Its core components include: a heterogeneous energy harvesting module cluster, an intelligent energy replenishment management center, a multi-input collaborative charging management module, an energy storage battery pack, and an online water quality monitoring instrument as the load. The selection, interfaces, and initialization configurations of each module are as follows:

[0032] Solar energy acquisition module configuration: The power generation unit uses a single monocrystalline silicon photovoltaic module with a peak power of 100W. Its open-circuit voltage (Voc) under standard test conditions (STC) is 24.5V, and its maximum power point voltage (Vmpp) is approximately 20.5V. The primary optimizer for maximum power point tracking uses a buck converter circuit built on a dedicated MPPT chip (e.g., TI's BQ24650). This circuit samples the photovoltaic module's port voltage V in real time. pv and current I pv The PWM duty cycle D is dynamically adjusted using a built-in incremental conductance algorithm, ensuring that the operating point always tracks the maximum power point. Its output is set to a stable DC bus voltage V. bus=14.6V (matching subsequent battery charging needs). The optimizer communicates with the intelligent power management hub via an isolated RS-485 interface. In the firmware, a structured energy characteristic tag is calculated and uploaded every 5 seconds, containing the following fields: Energy type coded as 0x01, representing solar energy. Rated output voltage: 14.6V.

[0033] Current instantaneous output power P solar : Calculated through internal sampling, unit: watt (W), P solar =V bus *I out , where I out The optimizer output current. Power fluctuation coefficient α solar This reflects short-term power stability. It is calculated by recording the P values ​​of the most recent 60 sampling points (i.e., the most recent 5 minutes). solar Given a sequence, calculate its standard deviation σ and mean μ, then α solar =σ / μ. If μ is close to 0, then α solar Setting it to a large constant (e.g., 10) indicates extreme instability. The expected sustainable power generation time T estsolar This is based on simple historical data extrapolation. For example, if the current time is 2 PM, then query the average light intensity at the same time (±30-minute window) over the past 3 days (which can be obtained via P). solar (E estimate after normalization) as E hist Currently P solar The converted light intensity is E now If E now >0.1*E hist (assuming there is effective lighting), then T estsolar Set the difference between the preset sunset time and the current time (in hours) multiplied by a decay factor γ (e.g., γ=E). now / E hist (and γ is between 0.3 and 1.0); if there is no effective illumination, then T estsolar Set it to 0.

[0034] The power generation unit in this invention employs a miniature horizontal-axis propeller-type hydroelectric generator, with an output power of approximately 10W at a rated flow velocity of 1.0m / s. The generator is a three-phase permanent magnet synchronous generator. The power optimizer is a module integrating a three-phase rectifier bridge, a BOOST boost circuit, and a digital controller. Its digital controller (using an STM32G031) runs a variable-step perturbation-observation (MPPT) algorithm suitable for low-speed water flow. This algorithm finds the point where maximum power is extracted from the rectified DC voltage by fine-tuning the duty cycle of the BOOST circuit. The optimizer output is also stabilized at V... bus=14.6V. Its energy characteristic tag is uploaded every 5 seconds via the UART interface: energy type code 0x02 represents water flow energy. Rated output voltage 14.6V. The current instantaneous output power P is measured and calculated. water Power fluctuation coefficient α water The calculation method is the same as α. solar Based on P water The historical sequence. Expected sustainable power generation time T estwater A simple Hall effect flow sensor is integrated into the module. Based on historical flow velocity v (m / s) data from the past hour, the average flow velocity v is calculated. avg Assuming the flow velocity changes steadily, the power generation potential over the next two hours can be estimated. A simplified calculation method is: if the current flow velocity v... now >0.3m / s (starting flow rate), then T estwater =2*(v now / v avg (hours), but not exceeding 6 hours in the future. If v now If the flow rate is lower than the startup flow rate, then T estwater =0. The power generation unit uses a piezoelectric-electromagnetic composite vibration energy harvester. Internally, it is a bistable cantilever beam with a PZT-5A piezoelectric ceramic plate fixed at one end and a magnet at the other end oscillating between two fixed coils under external vibration, simultaneously generating piezoelectric and induced electromotive force. To address the different output characteristics of piezoelectric (high voltage, high internal resistance) and electromagnetic (low voltage, low internal resistance), two independent AC-DC front-end processing circuits are used (the piezoelectric section uses full-bridge rectification + capacitor filtering, and the electromagnetic section uses full-bridge rectification + inductor filtering), which are then combined and input to a dedicated energy harvesting chip (such as the ADP5091). This chip integrates a boost converter and a maximum power point tracking circuit, efficiently harvesting irregular, milliwatt-level energy and boosting it to a usable DC voltage. In this design, its output is set at 5V, and then boosted to VV by a subsequent two-stage BOOST circuit. bus =14.6V. This module reports energy characteristic tags via the SPI interface: energy type code 0x03, representing vibration energy. Rated output voltage: 14.6V. Current instantaneous output power P vib : Obtained through measurement and calculation. Since vibration energy is typically an intermittent pulse, P vib Take the average power over the most recent 5 seconds.

[0035] Power fluctuation coefficient α vib Because vibrational energy is inherently highly volatile, α vib Use a fixed value of 2.0 or calculate dynamically based on the pulse duty cycle and amplitude. Expected sustainable power generation time T. estvib If N valid vibration events (power exceeding the threshold) are detected in the past 5 minutes eventIf the vibration rate is greater than 3 times per minute, it is considered to be continuous. (T) estvib Set to 1 hour; otherwise set to 0.

[0036] The energy storage battery used in this invention is a 12.8V / 100Ah battery pack composed of four 3.2V / 100Ah lithium iron phosphate (LiFePO4) cells connected in series. It is equipped with an intelligent battery management system with an SMBus communication interface, which can report the battery's state of charge (SOC) (0-100%), state of health (SOH) (capacity retention, 0-100%), temperature (T_bat), and the currently recommended charging stage (e.g., 0: constant current charging CC; 1: constant voltage charging CV; 2: float charging Float) in real time. The multi-input collaborative charging management module in this invention is a custom-designed power electronic circuit board. Its core is a reconfigurable switch matrix composed of high-performance MOSFETs (Infineon's OptiMOS series), controlled by an FPGA (such as Xilinx Spartan-6). This matrix can select the main charging path from the outputs of three optimizers: solar energy, hydroelectric energy, and vibration energy, by closing the corresponding switch (K). sol ,K wat ,K vib The system connects to the common main charging path. The auxiliary energy buffer and shaping path includes a shared buffer energy storage unit consisting of six supercapacitors (6 3000F / 2.7V capacitors connected in series, forming a capacitor bank of approximately 16.2V / 500F). When non-main charging energy needs temporary storage, it is connected via switch Kbufin to charge this capacitor bank. When it needs to be released, it is released through a Buck-Boost circuit (as a controllable current source) consisting of a MOSFET and an inductor with a controlled current Ibufout, controlled by switch Kbufout. The load direct compensation channel includes a MOSFET (K... bypass A bypass consisting of a DC-DC regulator and a DC-DC regulator. When enabled, K bypass The circuit closes, directly supplying the load with regulated electrical energy from the main supplementary energy source. The FPGA in this module receives instructions from the intelligent supplementary energy management center and executes specific switching actions and current control. The core of the intelligent supplementary energy management center is an ultra-low-power microcontroller (MCU) based on the ARM Cortex-M4 core, such as ST's STM32L4R9. It runs the FreeRTOS real-time operating system. The online monitoring instrument load is the water quality monitor in this embodiment. It has an RS-485 communication interface and can receive external commands to switch operating modes (such as normal mode, energy-saving mode 1, and energy-saving mode 2).

[0037] After the system powers on and completes initialization, the intelligent power replenishment management center begins to periodically execute the following process flow, controlling the cycle T.cycle Set to 5 seconds.

[0038] Step S1: Construct an independently optimizable heterogeneous energy harvesting module cluster. This step involves physical connection and continuous operation after system power-on. Each energy harvesting module operates independently and in parallel: the photovoltaic module generates DC power under illumination, and its MPPT primary optimizer continuously runs the incremental conductance algorithm to adjust the duty cycle D of the Buck circuit, ensuring the photovoltaic module operates near its maximum power point and outputs a stable 14.6V voltage to the system bus. Simultaneously, its internal MCU calculates P every 5 seconds. solar α solar and T estsolar The data is encapsulated into data frames and transmitted via RS-485. The hydroelectric generator produces three-phase AC power driven by the river. After three-phase rectification and BOOST MPPT circuitry in the integrated optimizer, it outputs 14.6V DC. Its digital controller calculates P every 5 seconds. water α water and T estwater The signal is transmitted via UART. The vibration energy harvester generates irregular alternating current under environmental vibration excitation. After two-way front-end processing by a composite optimizer and maximum power point tracking and boosting by the ADP5091 chip, a 5V output is obtained, which is then stabilized to 14.6V by a two-stage BOOST circuit. Its controller calculates P every 5 seconds. vib α vib and T estvib The outputs of these three modules are physically connected to the input of the multi-input collaborative charging management module, but they are initially decoupled from each other through their internal MPPT circuits to avoid energy mutual interference caused by direct parallel connection.

[0039] Step S2: Establish a multimodal collaborative sensing and decision-making network. In the MCU of the intelligent energy replenishment management center, a high-priority task (e.g., priority set to 3) Task_EnergyManager is triggered by a timer with a 5-second cycle. This task executes the following sub-steps sequentially:

[0040] S2.1 The MCU polls and reads the energy characteristic tag data of the solar energy acquisition module, water flow energy acquisition module, and vibration energy acquisition module through the corresponding communication interfaces (RS-485, UART, SPI). The MCU reads the battery's SOC, SOH, and T from the battery management system through the SMBus interface. bat Charge_Stage. The MCU reads the load current I through a high-precision sampling resistor and an ADC channel. load And calculate the current load power P. load = V bus *I load Vbus This is the system bus voltage (approximately 14.6V). Check that all data are within reasonable ranges (e.g., power is non-negative, SOC is between 0-100%). If a power source's data is invalid or has not been updated within a timeout period, mark that power source as unavailable and set its DPV (Dynamic Priority Value) to 0.

[0041] S2.2 For each available energy source i (i∈{solar,water,vib}), calculate its dynamic priority value DPV. i DPV i It is a dimensionless scalar value, intended to quantify the suitability of this energy source as a primary supplementary energy source under the current system conditions. This embodiment employs a weighted summation model: DPV i =w1*f normalize (P i )+w2*g stability (α i )+w3*h sustainability (Test i )+w4* k match (Charge_Stage, V_bus_i). The weighting coefficients w1, w2, w3, and w4 are set empirically, for example, w1=0.4, w2=0.2, w3=0.2, w4=0.2, and w1+w2+w3+w4=1. Each function is defined as follows: power factor f normalize (P i ): Normalize the power. f normalize (P i )=P i / P rated Among them, P rated This refers to the rated power of each energy source (e.g., 100W solar energy, 10W water flow energy, 1W vibration energy). This makes power of different magnitudes comparable. Stability factor g stability (α i ): Evaluate power fluctuations. g stability (α i ) = 1 / (1+α i ). α i The smaller the value, the more stable the function; the closer the function value is to 1. i The larger the value, the closer it approaches 0.

[0042] Sustainability factor h sustainability (Test i ): Evaluate the expected power generation duration. h sustainability (Test i ) = 1 - exp(-Test i / τ). Where τ is a time constant, for example, set to 2 hours. Test iThe longer the value, the closer it is to 1; Test i When k is 0, the value is 0. Matching factor k match (Charge_Stage,Vbus i Evaluation of energy output voltage Vbus i The degree of matching with the current charging stage of the battery. When Charge_Stage == 0 (constant current charging CC): the goal is to quickly replenish the battery's charge, and the voltage accuracy requirements are relatively relaxed, as long as Vbus... i Slightly higher than the current battery voltage is sufficient. Define the matching degree k. match =1.0 (if V_bus) i >V bat +0.5V), otherwise 0.5V. When Charge_Stage==1 (constant voltage charging CV): the goal is to precisely charge the battery with a constant voltage, requiring a very stable bus voltage. Define k match =1.0-β*|V_bus i -V CV |, where V CV V_bus represents the constant voltage charging setpoint of the battery (e.g., 14.6V for lithium iron phosphate), β is the proportional coefficient (e.g., 5.0), and |·| is the absolute value. i The closer to V CV k match The higher the value, the lower the required floating voltage. When Charge_Stage == 2 (Float): a lower floating voltage is required. Matching calculation is similar to the CV stage, but V... CV Replace with float charge voltage V float (e.g., 13.8V). Calculation example: Assume that at a certain moment, SOC=30%, and the battery is in the CC stage. Measure P. solar =25W,α solar =0.1,Test solar =4h; P water =8W, α water =0.05,Test water =2h;P vib =0.5W,α vib =1.5,Test vib =0h. V bus Both are 14.6V, meeting the CC stage requirements. Solar energy: f=25 / 100=0.25; g=1 / (1+0.1)=0.91; h=1-exp(-4 / 2)=1-exp(-2)=0.865; k=1.0. DPV solar=0.4*0.25+0.2*0.91+0.2*0.865+0.2*1.0=0.1+0.182+0.173+0.2=0.655 Water flow energy: f=8 / 10=0.8; g=1 / (1+0.05)=0.952; h=1-exp(-2 / 2)=0.632; k=1.0. DPV water =0.4*0.8+0.2*0.952+0.2*0.632+0.2*1.0=0.32+0.1904+0.1264+0.2=0.8368 Vibration Energy: Test vib =0, h=0, considered unreliable, DPV vib It can be counted as 0. Therefore, the DPV of the water flow energy is the highest in this cycle.

[0043] S2.3, Based on the calculated DPV i Based on battery status and load status, an arbitration decision is made to select the primary energy source, and the DPV of all available energy sources is compared. i The energy source with the highest value is selected as the primary energy source for the current cycle. In the example above, water flow energy is selected. The load direct compensation channel conditions are checked to determine whether load direct compensation should be enabled. The following conditions must be met simultaneously: a) The load is in a momentary high-power mode: Define P... load >Ploadavg*1.5, where Ploadavg is the average load power over the past minute. For example, if the average power consumption is 5W and the current instantaneous communication power reaches 15W, then it meets the requirement. b) Instantaneous output power P of the main supplementary power source. i Satisfy P i >P load *η, where η is the estimated efficiency of the direct compensation channel (e.g., 0.85). This means that the main compensation energy source is capable of directly supplying a portion of the load while charging the battery. c) The battery SOC is higher than a safety threshold (e.g., 20%) to prevent unstable direct compensation when the battery is low. If the condition is met, a strategy including the load direct compensation channel mode is generated. Otherwise, this mode is not enabled. Define the auxiliary compensation energy source and its processing method. Other available energy sources not selected as the main compensation energy source, if their DPV i If the output voltage is greater than a minimum threshold (e.g., 0.1), it is defined as an auxiliary power source. For auxiliary power sources, their output voltage Vbus is further determined. i The degree of matching with the bus voltage required for current battery charging. If |Vbus i -V targetIf the voltage exceeds ΔV (ΔV is a set tolerance, such as 0.5V), it is considered a mismatch, and its electrical energy will be guided to the shared buffer energy storage unit for temporary storage and shaping. If it matches, it can serve as a potential secondary charging source, but in this embodiment, for the sake of simplifying control, the buffer unit strategy is preferred. Finally, the decision-maker generates a structured energy dispatching strategy instruction, including: {primary energy source ID, whether to enable direct load compensation, auxiliary energy source 1}. ID Its processing method (direct charging / buffering), auxiliary energy replenishment 2 ID And its processing methods}.

[0044] Step S3: Perform charging management and energy distribution based on multi-input collaboration.

[0045] The FPGA of the multi-input collaborative charging management module receives energy scheduling strategy instructions from the MCU, parses them, and executes the corresponding hardware control operations.

[0046] S3.1. Assume the current strategy instructions are: {Main supplementary energy: water flow energy, enable direct load supplementation: Yes, auxiliary supplementary energy 1 (solar energy): Buffer, auxiliary supplementary energy 2 (vibration energy): Ignore}. The main path controls the FPGA control switch matrix, which is closed-loop connected to the switch K output by the water flow energy optimizer. wat Disconnect K sol and K vib At this point, the electrical energy generated by the water flow flows to the subsequent battery charging management circuit through the low-impedance main charging path. Because load direct compensation is enabled, the FPGA simultaneously closes the bypass switch K. bypass Before entering the battery charging management circuit, a portion of the electrical energy from the water flow is diverted and, after passing through a high-efficiency DC-DC step-down regulator (e.g., from 14.6V to 12V), directly supplies the water quality monitoring instrument load. The diversion ratio can be roughly controlled by adjusting the current in the branch containing K_bypass or through current feedback in the main path, with the goal of prioritizing the instantaneous peak power demand of the load, and then using any surplus energy to charge the battery. For solar energy (auxiliary energy source 1, determined to be mismatched and requiring buffering), the FPGA closes the switch Kbuf_in (connected to the output of the solar energy optimizer). The solar energy is then fed into a shared buffer energy storage unit (supercapacitor bank). The voltage V of the supercapacitor... cap It begins to rise. FPGA monitoring V cap When it reaches its upper limit (e.g., 15.5V) or solar power input stops, Kbuf_in is disconnected. At the same time, a buffered energy release management subroutine is triggered.

[0047] S3.2 The working logic of the above buffer energy release management subroutine is as follows: Release timing judgment: when the instantaneous power P of the main supplementary energy... mainWhen the current charging current is insufficient to meet the battery's current setpoint Ichgset, there exists a power difference ΔP = Ichgset * V. bat -P main (V) bat (This refers to the current battery voltage). Furthermore, the voltage V of the shared buffer energy storage unit... cap Higher than battery voltage V bat A minimum margin (e.g., 0.7V). Release control: If the release timing is met, the FPGA closes the switch Kbuf_out and controls the subsequent Buck-Boost circuit to operate in controllable current source mode. Its output current Ibuf_out is set such that Ibuf_out ≈ ΔP / V. bat In this way, the current Ibuf_out from the buffer unit and the current I_main (≈P_main / V_bat) from the main power source are synchronized in time, and the currents are added at the circuit nodes to jointly provide the charging current I for the battery. chg =I main +Ibuf_out thus smoothly and efficiently utilizes solar energy that would otherwise be wasted due to mismatch.

[0048] Step S4: Implement adaptive closed-loop management of battery and load

[0049] This step runs in parallel with S2 / S3, or can be performed in the intervals between them, and mainly consists of two asynchronous tasks:

[0050] S4.1, A low-priority task, Task_EnergyForecast, executes once per minute: Future energy balance prediction; the task reads the latest Test... solar Test water Test vib Assume the evaluation period is T. period Set for the next 2 hours. For each energy source i, its projected available energy E during this assessment period. i It can be simply estimated as: E i =Pi avg *min(Test i ,T period ), where Pi avg This is the average power of the energy source over the past 15 minutes. Total energy prediction: E_total_predicted = ΣE i(Sum of all available energy sources). Load demand forecast: The energy consumption of the load in the next 2 hours, E_load_required, can be estimated based on its current operating mode and historical power consumption. For example, if maintaining the current normal mode (average 5W), then E_load_required = 5W * 2h = 10Wh. The amount of energy needed to replenish the battery, E_bat_needed, also needs to be considered, depending on the target SOC (e.g., 80%) and the current SOC. If the forecast satisfies (E_total_predicted * η...), then... sys )<(E_load_required+E_bat_needed), where η sy If 's' represents the system's estimated efficiency (e.g., 0.7), then it is assumed that there will be insufficient energy in the future. The intelligent energy replenishment management center sends a negotiation message to the water quality monitor via the RS-485 interface, suggesting, for example, switching to energy-saving mode 1. The message includes the estimated energy shortage and its duration. Upon receiving the message, the monitor can decide whether and when to switch modes based on preset strategies (e.g., reducing sensor sampling frequency, extending data reporting intervals), and confirms via communication, thus achieving closed-loop management of the energy budget.

[0051] S4.2 Battery health status (SOH) directly affects the safety boundary of the charging strategy. The SOH value reported by the battery management system (e.g., 85%) is used to calculate the matching factor k of the dynamic priority value DPV in the intelligent charging management center. match This introduces the influence of State of Health (SOH). For example, for batteries with decreased health (SOH < 90%), the maximum allowable charging current I_chg_max can be appropriately reduced during the constant current charging phase. Correspondingly, the I_chg_set value used when determining the load compensation channel conditions and the timing of buffer energy release will also decrease based on SOH. In the DPV calculation model, the weight w2 (stability factor) can be dynamically adjusted. For batteries with low SOH, the weight of w2 can be appropriately increased, making the system more inclined to select batteries with smaller power fluctuations (α). i Smaller energy sources are used as the primary supplementary energy source to reduce impact on the battery and achieve protective charging. Throughout the system operation, steps S1 to S4 form a closed-loop, adaptive intelligent energy replenishment process: S1 is the continuous supply of underlying data. S2 is a decision-making process every 5 seconds, integrating all current information (ambient energy, battery status, load status) to calculate the optimal energy utilization strategy. S3 is a rapid actuator that precisely controls the power switching network according to the instructions from S2, achieving efficient energy distribution and convergence. S4 is a slower health and budget management step, optimizing from a longer timescale and system lifespan perspective.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for online monitoring of green energy supplementation, characterized in that, This method is implemented based on a power replenishment system, which includes a heterogeneous energy harvesting module cluster, an intelligent power replenishment management center, a multi-input collaborative charging management module, an energy storage battery, and an online monitoring instrument load. The method specifically includes the following steps: S1. Construct a cluster of heterogeneous energy harvesting modules that can be independently optimized. For three types of energy sources, namely solar energy, hydroelectric energy, and vibration energy, energy harvesting and preliminary stabilization output are carried out through independent power generation units and their integrated maximum power point tracking primary optimizers. A unique energy characteristic label is defined for each type of energy. S2. Establish a multimodal collaborative perception and decision-making network. The intelligent energy replenishment management center performs real-time data acquisition, dynamic priority evaluation, arbitration and strategy generation to determine the main energy source, auxiliary energy source and their working mode in the current cycle. S3. Execute charging management and energy distribution based on multi-input collaboration. The multi-input collaborative charging management module performs reconfigurable path management and intelligent convergence of electrical energy from the main energy source and auxiliary energy source according to the strategy instructions generated by the intelligent energy replenishment management center, so as to charge the energy storage battery. S4. Implement adaptive closed-loop management of battery and load, including adaptive load negotiation based on energy prediction and battery health-oriented charging based on battery health status.

2. The online monitoring method for green energy replenishment according to claim 1, characterized in that, In step S1, the energy characteristic label includes at least: energy type, rated output voltage range, current instantaneous output power, power fluctuation coefficient, and expected sustainable power generation time based on historical data.

3. The online monitoring method for green energy replenishment according to claim 1, characterized in that, The dynamic priority assessment in step S2 specifically involves the intelligent energy replenishment management center calculating a dynamic priority value for each available energy source. The calculation of this dynamic priority value comprehensively considers the instantaneous output power level, power stability, expected sustainable power generation time, and the degree of matching with the current charging stage requirements of the energy storage battery.

4. The online monitoring method for green energy replenishment according to claim 3, characterized in that, The arbitration and strategy generation in step S2 further includes: when it is determined that the online monitoring instrument load is in an instantaneous high power consumption mode, and the dynamic priority value and output power of the main energy source meet the preset conditions, a strategy including the load direct compensation channel mode is generated, and the instruction is given to divert part of the electrical energy generated by the energy source to directly supply the online monitoring instrument load.

5. The online monitoring method for green energy replenishment according to claim 1, characterized in that, The reconfigurable pathway management of electrical energy from the auxiliary power source in step S3 includes: if the output characteristics of the auxiliary power source do not match the current battery charging demand, then guide it to a shared buffer energy storage unit for temporary storage and shaping of electrical energy.

6. The online monitoring method for green energy replenishment according to claim 5, characterized in that, The intelligent current convergence in step S3 specifically refers to the following: the power path for future autonomous energy replenishment and the power released from the shared buffer energy storage unit are interleaved in time or phase to form a charging current waveform for the energy storage battery.

7. The online monitoring method for green energy replenishment according to claim 1, characterized in that, The adaptive load negotiation based on energy prediction in step S4 specifically includes: the intelligent energy replenishment management center predicts, based on the expected sustainable power generation time of each energy source and historical energy collection data, that the total energy available to replenish the energy storage battery will be lower than the energy required to maintain the online monitoring instrument load at rated power consumption within the next preset evaluation period, and then sends an early warning to the online monitoring instrument load and negotiates to enter the graded energy saving mode.

8. The online monitoring method for green energy replenishment according to claim 1, characterized in that, The battery health-oriented charging based on battery health status in step S4 specifically includes: using the health status of the energy storage battery as one of the input parameters for calculating the dynamic priority value, and adopting a charging strategy that reduces the charging current threshold and avoids using energy with drastic fluctuations as the main supplementary energy for batteries with declining health status.

9. The online monitoring method for green energy replenishment according to claim 1, characterized in that, The heterogeneous energy harvesting module cluster includes a solar energy harvesting module, a water flow energy harvesting module, and a vibration energy harvesting module. The solar energy harvesting module uses a monocrystalline silicon photovoltaic panel and its maximum power point tracking DC-DC optimizer. The water flow energy harvesting module adopts an integrated design of a micro turbine generator and a power optimizer. The vibration energy harvesting module adopts a piezoelectric electromagnetic composite energy harvester.

10. The method for online monitoring of green energy replenishment according to claim 1, characterized in that, The multi-input collaborative charging management module contains a reconfigurable network composed of high-speed power switches, which is used to switch power paths according to strategy instructions; the intelligent power replenishment management hub is implemented by an ultra-low power microcontroller, which is used to run a dynamic priority evaluation algorithm.