Cathodic protection power supply device based on photovoltaic power supply and edge calculation and control method

Abstract

The application provides a cathode protection power supply device and a control method based on photovoltaic power supply and edge computing, constructs an energy management structure of photovoltaic priority and battery buffer, directly supplies power by photovoltaic when photovoltaic output meets the demand of cathode protection electric energy, and stores the remaining electric energy in an energy storage unit; when the photovoltaic output is insufficient, automatically switches to supply power by the energy storage unit, reduces double energy conversion loss, reduces the frequency of battery charging and discharging, prolongs the service life of energy storage, and improves the stability of the output potential of the cathode protection power supply device; at the same time, an edge computing unit is arranged in the device, and a multi-source data acquisition interface and a self-identification pre-adjustment module are integrated side by side, local fusion, rapid calculation and parameter correction are realized, and the device can stably operate even if the network is disconnected; an interference suppression and intelligent hierarchical early warning integrated execution module is arranged, hardware filtering and dynamic compensation are adopted, and risk level determination and hierarchical warning are completed locally, and the anti-interference stability and risk visualization effect are improved.

Description

Technical Field

[0001] This invention relates to the field of power supply technology, and in particular to a cathodic protection power supply device and control method based on photovoltaic power supply and edge computing. Background Technology

[0002] In recent years, distributed energy technology has continued to develop in the fields of photovoltaic power generation, energy storage systems, and intelligent control. Improvements in photovoltaic conversion efficiency, reductions in energy storage costs, and advancements in control technology have enhanced the power supply stability of independent small-scale power supply systems and reduced operation and maintenance costs, making them a viable distributed energy supply method as the core power source (potential constant meter) for forced current cathodic protection of oil and gas pipelines. In areas where mains power is unavailable or supply is unstable, such as pipeline sections in communities far from substations or areas with DC interference from subways, traditional long-distance power transmission schemes are complex to construct, costly, and easily subject to environmental constraints. In contrast, localized energy supply schemes based on distributed energy can directly collect and utilize clean energy on-site, avoiding the losses and safety hazards associated with long-distance power transmission.

[0003] In existing technologies, distributed power supply cathodic protection devices typically employ two types of data processing methods. One type of device is equipped with a local storage unit to periodically record the reference potential at the junction point, environmental parameters, and output current. The data can be manually exported for analysis, exhibiting low network dependence, but it cannot respond promptly to external interference or changes in operating status. The other type of device supports real-time or periodic uploading of data to the cloud via communication methods such as 4G, NB-IoT, or LoRaWAN, facilitating centralized management and trend analysis. However, its response speed is limited under short-term interference due to network latency and interruptions.

[0004] Regardless of the data processing mode employed, existing equipment mostly uses a fixed "photovoltaic-battery-load" series power supply structure. This structure typically inputs photovoltaic power into an energy storage unit, which then powers the load. This secondary energy conversion reduces utilization efficiency, and frequent battery charging and discharging affects battery lifespan. Furthermore, existing equipment has limited local computing and control capabilities, making it unable to respond promptly to random changes in pipeline protection requirements. Especially in environments with strong interference such as rail AC / DC interference, ground current interference, and high-voltage DC grounding electrodes, the filtering and compensation processes are prone to lag, potentially causing short-term protection failures and increasing the risk of pipeline corrosion. Summary of the Invention

[0005] The present invention aims to provide a cathode protection power supply device and control method based on photovoltaic power supply and edge computing control that overcomes or at least partially solves the above problems.

[0006] To achieve the above objectives, the technical solution of the present invention is specifically implemented as follows:

[0007] One aspect of the present invention provides a cathodic protection power supply device based on photovoltaic power supply and edge computing control, comprising: a photovoltaic module, a controller, an energy management unit, an edge computing unit, an execution unit, an energy storage unit, and a power output device; wherein:

[0008] The photovoltaic module is electrically connected to the energy management unit and is used to convert solar energy into direct current (DC) power and provide the DC power to the energy management unit.

[0009] An energy management unit, electrically connected to the energy storage unit and the controller, is used to interact with the energy storage unit, charge and discharge the energy storage unit, and distribute energy under the scheduling of the controller;

[0010] The controller is used to collect the output voltage and current of the photovoltaic module, the voltage and current of the energy storage unit, and the load demand of the cathodic protection power supply. When the photovoltaic output power is higher than the load demand of the cathodic protection power supply and the voltage of the energy storage unit has not reached the upper limit threshold, the controller controls the energy management unit to use the DC power converted by the photovoltaic module to supply power and charge the energy storage unit. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is greater than the lower limit threshold, the controller controls the energy management unit to disconnect the DC power supply path converted by the photovoltaic module and use the energy storage unit to supply power. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is less than or equal to the lower limit threshold, a low power alarm signal is generated. The controller also receives adjustment instructions from the edge computing unit to maintain the output potential of the cathodic protection power supply device at no lower than the set protection lower limit potential.

[0011] The edge computing unit is used to collect protection potential, output current, output voltage, environmental parameters and pipeline interference signals in real time, perform local fusion calculations based on the collected data, predict the potential deviation trend in advance, and generate the adjustment command to be sent to the controller or the execution unit.

[0012] The execution unit is used to receive the adjustment command, perform hardware filtering and dynamic compensation on the output control signal, and apply the compensated control quantity to the power output device; at the same time, it classifies the risk level according to the intensity and duration of the interference and triggers an alarm.

[0013] The power output device is used to provide the power output signal, which has been adjusted by the execution unit, to the external cathodic protection circuit, serving as the power output interface of this device.

[0014] Optionally, the edge computing unit performs local fusion calculations based on the collected data to predict the potential deviation trend in advance and generate adjustment commands in the following manner:

[0015] Set the target protection interval [E] min E max Safety upper limit, safety lower limit, power and thermal constraints;

[0016] The collected data was subjected to first-order exponential smoothing and median filtering for noise reduction.

[0017] off (k)=λ off (k-1)+(1-λ) E off (k), λ∈[0.7,0.95]

[0018] in, off (k) is the smoothed power-off potential of the kth sample, which is the denoised estimate obtained by first-order exponential smoothing filter, and λ is the smoothing coefficient.

[0019] Calculate instantaneous deviation:

[0020] e(k) = E set- off (k)

[0021] Where e(k) is the potential deviation of the k-th sampling, E set Set the protection potential for the system;

[0022] Calculate the rate of change of deviation:

[0023] Δe(k) = e(k) - e(k-1)

[0024] Calculate trend strength:

[0025] Δe EWMA (k)=β Δe(k)+(1-β) Δe EWMA (k-1), β∈[0.3,0.7]

[0026] Where, Δe EWMA (k) represents the rate of change of the exponentially weighted moving average deviation in the kth iteration, and β is the smoothing weight factor;

[0027] The PID algorithm is used for online adjustment of control parameters:

[0028] u(k)=K p [e(k)-e(k-1)]+K i e(k)+K d [e(k)-2e(k-1)+e(k-2)]

[0029] Where u(k) is the controller output, the equivalent control input for directly driving the DC / DC or cathodic protection power supply, e(k) is the potential deviation of the k-th sample, and K p K is the proportionality coefficient. i K is the integral coefficient. d These are the differential coefficients;

[0030] By adding a trend feedforward and short-term prediction term before the PID output, we get ê(k+1);

[0031] Calculate the pre-adjustment feedforward:

[0032] u ff (k) = αê(k+1), α > 0

[0033] Among them, u ff (k) is the trend feedforward control quantity, and α is the feedforward gain coefficient;

[0034] Perform synthesis control:

[0035] u pre (k)=u ff (k)+u(k)

[0036] Among them, u pre (k) is the pre-adjusted synthetic control signal;

[0037] For u pre (k) Apply soft / hard limiting and slope constraints:

[0038] u min ≤u pre (k)≤u max ,∣u pre (k)-u pre (k-1)∣≤r max

[0039] Among them, u min u is the lower limit value of the output control signal. max r is the upper limit of the output control signal. max This represents the upper limit of the output rate of change.

[0040] Optionally, the edge computing unit uses a PID algorithm to adjust the control parameters online in the following manner:

[0041] The discrete form and parameter tuning rules include: using the potential deviation value e and the rate of change of potential deviation Δe as inputs, adjusting the proportional coefficient K online. p Integral coefficient K i and differential coefficient K d ;

[0042] When |e|>0.1V, increase the proportionality coefficient K. p Decrease the differential coefficient K d To quickly eliminate bias, and then reduce the integral coefficient K after stabilization. i Suppress overshoot;

[0043] When |e| < 0.1 V, decrease the proportionality coefficient K. p Differential coefficient K d Improve steady-state accuracy.

[0044] Optionally, the edge computing unit adds a trend feedforward and short-term prediction term before the PID output in the following manner:

[0045] Using moving linear prediction:

[0046] ê(k+1) =ae(k)+(1-a)e(k-1), a∈[0.6,0.9]

[0047] Where ê(k+1) is the predicted potential deviation value at time k+1, and a is the moving linear prediction weight coefficient;

[0048] Or with Δe EWMA Take a forward look:

[0049] ê(k+1)=e(k)+Δe EWMA (k).

[0050] Optionally, the controller is further configured to appropriately increase the differential coefficient K when a significant fundamental frequency component or diurnal energy peak is detected in the potential signal. d Reduce the integral coefficient K i And reduce the feedforward gain α.

[0051] Optionally, the controller is further configured to temporarily increase the proportional coefficient K when the rate of change of the potential deviation |Δe| exceeds a preset threshold. p And increase the feedforward gain α.

[0052] Optionally, the execution unit receives the adjustment command in the following manner, performs hardware filtering and dynamic compensation on the output control signal, and applies the compensated control quantity to the power output device:

[0053] The interference is decomposed into interference components in different frequency bands, such as quasi-DC / slow drift, low-frequency oscillation, power frequency and harmonics, and the corresponding dynamic compensation amount is output for each.

[0054] The dynamic compensation is superimposed on the output signal of the power supply control circuit and applied directly to the power output device; or, the dynamic compensation is injected into the equivalent setting port or PWM duty cycle control port of the power output device.

[0055] Optionally, the execution unit outputs the DC offset compensation amount in the following manner:

[0056] b(k+1)=b(k)+w(k), w~N(0,Q)

[0057] E dc (k)=E true (k)+b(k)+v(k), v~N(0,R)

[0058] Kalman filtering or exponential smoothing methods are used to address slow drift. (k) Estimate the offset to obtain the estimated value and generate the corresponding DC compensation amount:

[0059] u dc (k)=K dc (k), 0 <K dc ≤1

[0060] Where b(k+1) is the slow drift or zero bias estimate term calculated in the (k+1)th time, w(k) is the process noise term, and E dc (k) represents the DC potential measurement value from the kth sampling, E true (k) represents the true DC potential, v(k) represents the measurement noise term, and u dc (k) is the DC compensation control quantity, K dc This is the DC compensation gain coefficient.

[0061] Optionally, the execution unit outputs the AC interference compensation amount in the following manner:

[0062] The disturbance is approximated as a superposition of sinusoidal bases:

[0063]

[0064] ω m Tracked by PLL / FLL; LMS updates minimize residuals:

[0065] ε(k)=E ac (k)- ,

[0066] Calculate the compensation amount:

[0067]

[0068] in, E is the estimated value of the AC interference signal from the k-th sample; ac (k) represents the measured AC potential signal from the kth sampling. Let be the orthogonal basis vector of the m-th interference component; The angular frequency of the m-th interference component is determined by PLL / FLL tracking; T s The sampling period; w m (k) : These are the weight coefficient vectors of the m-th component and their estimated values, respectively; The step size coefficient of the LMS algorithm controls the weight update speed and convergence stability; This is the residual signal, which is the difference between the actual measured signal and the estimated signal; The amount of compensation for AC interference is the negative value of the estimated interference to achieve cancellation.

[0069] Optionally, the execution unit outputs the periodic disturbance compensation amount in the following manner:

[0070] Periodic disturbance compensation is adopted:

[0071] u rep (k)=u rep (kN)+Q(z) e(kN),

[0072] Where Q(z) is a low-pass stabilization filter, u rep (k) is the periodic disturbance compensation amount, e(k) is the potential deviation, and T period For the interference period, T s The sampling period.

[0073] Optionally, the execution unit outputs the feedforward compensation amount in the following manner:

[0074] Reference feedforward compensation is adopted:

[0075] u ff (k) = G ff (z) d ref (k)

[0076] Among them, G ff (z) is the feedforward transfer function determined by calibration or identification, d ref (k) is the reference disturbance signal, u ff (k) represents the reference feedforward compensation amount.

[0077] Optionally, the execution unit is also configured to synthesize compensation and limit amplitude / slope:

[0078] u raw (k)=u dc (k)+u ac (k)+u rep (k)+u ff (k)

[0079] u comp (k)=clip (u raw (k), u min ,u max ),∣u comp (k)-u comp (k-1)∣≤r max

[0080] Among them, u raw (k) represents the original synthesized compensation signal without amplitude limiting processing; u comp (k) is the dynamically compensated synthesized signal output by the execution unit; u dc (k) represents the DC offset compensation amount; u ac (k) represents the AC interference compensation amount; u rep (k) represents the periodic disturbance compensation amount; u ff (k) represents the trend feedforward compensation amount; u min u max These are the lower and upper limits of the output limiting, respectively; r max This represents the upper limit of the output rate of change.

[0081] When predicting off If the gain is about to exceed the limit, the adaptive gain is temporarily frozen, and a minimum compensation is applied while the recovery is gradually released.

[0082] Another aspect of the present invention provides a control method for a cathodic protection power supply device based on photovoltaic power supply and edge computing, comprising:

[0083] Collects photovoltaic, energy storage, potential, current, voltage, and interference signals;

[0084] Implement a photovoltaic-priority energy management strategy;

[0085] Local denoising, feature extraction, and trend prediction are performed on potential and interference signals.

[0086] Adjustment instructions are generated based on deviation, rate of change, and trend.

[0087] Performs DC compensation, AC compensation, periodic disturbance compensation, and feedforward compensation;

[0088] Synthesize the compensation amount and apply amplitude and slope limiting to the output;

[0089] The risk level is determined and an early warning is issued based on the intensity and duration of the interference.

[0090] Therefore, the cathode protection power supply device and control method based on photovoltaic power supply and edge computing provided by this invention constructs an energy management structure that prioritizes photovoltaic power and buffers battery power. When the photovoltaic output meets the power consumption requirements of cathode protection, the photovoltaic power supply is directly supplied, and the remaining power is stored in the energy storage unit. When the photovoltaic output is insufficient, the power supply is automatically switched to the energy storage unit, reducing the double energy conversion loss, lowering the battery charging and discharging frequency, extending the energy storage life, and improving the stability of the output potential of the cathode protection power supply device. At the same time, the device is equipped with an edge computing unit, which integrates multi-source data acquisition interfaces and self-identification pre-adjustment modules in parallel to achieve local fusion, fast calculation and parameter correction, and can operate stably even when the network is disconnected. An integrated execution module for interference suppression and intelligent hierarchical early warning is set up, which uses hardware filtering and dynamic compensation, and completes risk level determination and hierarchical alarm locally, improving anti-interference stability and risk visualization. Attached Figure Description

[0091] 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.

[0092] Figure 1 This is a schematic diagram of the structure of a cathode protection power supply device based on photovoltaic power supply and edge computing control provided in an embodiment of the present invention;

[0093] Figure 2 A flowchart illustrating the control method for a cathode protection power supply device based on photovoltaic power supply and edge computing, provided in an embodiment of the present invention. Detailed Implementation

[0094] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0095] The core of the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in this embodiment of the invention lies in:

[0096] 1. Energy management architecture prioritizing photovoltaic power and buffering with battery storage

[0097] When the photovoltaic output meets the power consumption requirements of cathodic protection, the system is directly powered by the photovoltaic system, and the remaining energy is stored in the energy storage unit. When the photovoltaic output is insufficient, the system automatically switches to power supply from the energy storage unit, avoiding secondary energy conversion, reducing battery charging and discharging frequency, and extending battery life. The system includes a solar power module (integrating an MPPT controller and lightning surge protection circuit), a lithium battery pack (BMS protection), and a high-efficiency DC-DC boost module (0–50V / 30A, efficiency ≥92%).

[0098] 2. Fast local control with integrated edge computing

[0099] With a built-in edge computing module, it receives protection potential and interference signal data in real time, and performs fusion calculation and rapid adjustment locally; even when the network is interrupted or interference fluctuates, it can still ensure low-latency potential adjustment and local fault diagnosis.

[0100] 3. Integrated design of anti-interference and dynamic compensation

[0101] The integrated hardware filtering and dynamic compensation algorithm can maintain stable output potential under the interference conditions of rail AC DC, high voltage DC grounding electrode and stray ground current; the interference detection and suppression module is parallel to the constant potential control loop, shortening the compensation response link and improving anti-interference capability.

[0102] 4. Intelligent hierarchical early warning and local decision-making functions

[0103] After interference detection, the system completes the risk level determination locally (such as low, medium and high levels) and triggers corresponding protection strategies or alarm methods according to the level; the early warning information can be reported to the cloud through the communication module to realize remote monitoring and operation and maintenance decision-making; the graded strategy can be dynamically adjusted according to different interference types such as rail AC DC, high voltage DC, and ground current, improving the system's adaptability and visualization level in complex electromagnetic environments.

[0104] 5. Combining local real-time response with long-term cloud-based management

[0105] Compatible with both local storage and IoT cloud-based data processing modes: In local storage mode, the device can operate independently and export data periodically, resisting network dependence; in cloud mode, it supports real-time data reporting, trend analysis, and centralized management via 4G / NB-IoT. The system can maintain constant potential protection when offline, and achieve long-term trend analysis and remote parameter optimization when connected to the network.

[0106] The cathode protection power supply device based on photovoltaic power supply and edge computing control provided by the present invention includes:

[0107] Photovoltaic modules are used to convert solar energy into direct current (DC) power and provide the DC power to an energy management unit.

[0108] The energy management unit is used to interact with the energy storage unit, charge and discharge the energy storage unit, and execute energy distribution strategies such as direct photovoltaic power supply, photovoltaic charging and energy storage discharge under the scheduling of the controller.

[0109] The controller is used to collect real-time operating parameters of photovoltaic modules, energy storage units and cathodic protection power supply loads, and execute a photovoltaic-first energy management strategy based on photovoltaic power, energy storage status and load demand; at the same time, it receives adjustment instructions from the edge computing unit to make the cathodic protection power supply device work within the set protection potential range.

[0110] The edge computing unit is used to collect protection potential, output current, output voltage, environmental parameters and pipeline interference signals in real time. Based on local computing, it performs data preprocessing, deviation calculation, trend analysis and interference identification, and generates dynamic adjustment commands to be sent to the controller or execution unit.

[0111] The execution unit is used to receive adjustment commands, adjust the output of the cathodic protection power supply in real time through hardware filtering, dynamic compensation and risk classification strategies, and classify the risk level according to the interference intensity, rate of change and duration and trigger alarms.

[0112] The power output device is used to provide the power output signal regulated by the execution unit to the external cathodic protection circuit, serving as the power output interface of this device and establishing an electrical connection with the external protected structure.

[0113] Optionally, the energy management unit executes the photovoltaic priority strategy in the following manner:

[0114] When the photovoltaic output power is higher than the load requirement of the cathode protection power supply and the voltage of the energy storage unit has not reached the upper limit threshold, the photovoltaic direct power supply is driven and the energy storage unit is charged simultaneously.

[0115] When photovoltaic power is insufficient and the voltage of the energy storage unit is higher than the lower threshold, the system switches to energy storage power supply.

[0116] Optionally, the edge computing unit processes data in the following manner:

[0117] The acquired protection potential and interference signals are subjected to denoising processing, including first-order exponential smoothing, median filtering and feature extraction;

[0118] Key characteristic quantities such as potential deviation, deviation change rate, and trend strength are calculated based on the preprocessing results.

[0119] The parameters of the fuzzy adaptive PID are adjusted dynamically based on the deviation and its changing trend.

[0120] Short-term trend prediction is performed to generate feedforward quantities, thereby improving the response speed and stability of power output regulation.

[0121] The execution unit performs dynamic compensation in the following manner:

[0122] The interference is decomposed into interference components in different frequency bands, such as quasi-DC drift, low-frequency oscillation, power frequency and harmonics, and periodic disturbances.

[0123] The DC compensation is generated by using Kalman filtering or exponential smoothing to align the DC component.

[0124] The fundamental frequency and phase of the AC interference components are dynamically tracked by PLL / FLL, and the residual is minimized by the sinusoidal LMS adaptive algorithm to obtain the AC interference compensation amount.

[0125] Repetitive control is used to generate periodic compensation quantities for periodic disturbances;

[0126] The compensation components are synthesized, and the synthesized compensation is then constrained by amplitude and slope limits before being injected into the power output regulation channel.

[0127] Optionally, the execution unit is further configured to freeze the adaptive gain and perform a safety margin compensation when the predicted potential goes out of bounds, in order to prevent overshoot and oscillation.

[0128] The following provides a detailed description of the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in the embodiments of the present invention: Figure 1 This diagram illustrates the structure of a cathode protection power supply device based on photovoltaic power supply and edge computing control provided in an embodiment of the present invention. (See also...) Figure 1 The cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in this embodiment of the invention includes: a photovoltaic module, a controller, an energy management unit, an edge computing unit, an execution unit, an energy storage unit, and a power output device; wherein:

[0129] The photovoltaic module is electrically connected to the energy management unit and is used to convert solar energy into direct current (DC) power and provide the DC power to the energy management unit.

[0130] An energy management unit, electrically connected to the energy storage unit and the controller, is used to interact with the energy storage unit, charge and discharge the energy storage unit, and distribute energy under the scheduling of the controller;

[0131] The controller is used to collect the output voltage and current of the photovoltaic module, the voltage and current of the energy storage unit, and the load demand of the cathodic protection power supply. When the photovoltaic output power is higher than the load demand of the cathodic protection power supply and the voltage of the energy storage unit has not reached the upper limit threshold, the controller controls the energy management unit to use the DC power converted by the photovoltaic module to supply power and charge the energy storage unit. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is greater than the lower limit threshold, the controller controls the energy management unit to disconnect the DC power supply path converted by the photovoltaic module and use the energy storage unit to supply power. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is less than or equal to the lower limit threshold, a low power alarm signal is generated. The controller also receives adjustment instructions from the edge computing unit to maintain the output potential of the cathodic protection power supply device at no lower than the set protection lower limit potential.

[0132] The edge computing unit is used to collect protection potential, output current, output voltage, environmental parameters and pipeline interference signals in real time, perform local fusion calculations based on the collected data, predict the potential deviation trend in advance, and generate the adjustment command to be sent to the controller or the execution unit.

[0133] The execution unit is used to receive the adjustment command, perform hardware filtering and dynamic compensation on the output control signal, and apply the compensated control quantity to the power output device; at the same time, it classifies the risk level according to the intensity and duration of the interference and triggers an alarm.

[0134] The power output device is used to provide the power output signal, which has been regulated by the execution unit, to the external cathodic protection circuit.

[0135] Specifically, the overall structure of the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in this embodiment of the invention may include photovoltaic modules, energy storage units, energy management units, controllers, edge computing units, execution units, and power output devices. Each unit works collaboratively through electrical connections and data interaction. Wherein:

[0136] 1. Photovoltaic modules can utilize solar power supply modules and incorporate a built-in Maximum Power Point Tracking (MPPT) controller and lightning surge protection circuit. Under sunlight conditions, photovoltaic modules can efficiently convert solar energy into DC power and maintain the output power at the optimal level through the MPPT algorithm; their protection circuit effectively prevents damage to the device from lightning strikes and surges. The output of the photovoltaic modules can directly supply power to the controller, ensuring "photovoltaic priority" power supply when photovoltaics are abundant; on the other hand, it can interact with the energy management unit and energy storage unit to achieve charging and feedback of electrical energy.

[0137] 2. The energy storage unit can be composed of lithium battery packs and equipped with a battery management system (BMS), providing overcharge, over-discharge, over- or under-temperature protection, and short-circuit protection. In situations of insufficient photovoltaic power, the energy storage unit can provide a stable power supply to the power output device; in situations of excess photovoltaic power, the energy storage unit can absorb excess energy, achieving dynamic energy buffering. The energy storage unit can be connected to the photovoltaic modules through an energy management unit and to the cathode protection power supply device through a controller, enabling bidirectional energy flow.

[0138] 3. The energy management unit may include an energy switching control module and a bidirectional DC / DC conversion module. The energy switching control module is responsible for switching between three modes: direct photovoltaic drive, energy storage charging, and energy storage discharging. The DC / DC conversion module is used to regulate the voltage to ensure energy transfer efficiency and voltage matching between different units. As a bridge between the photovoltaic modules and the energy storage units, the energy management unit achieves optimal energy allocation under the control of the controller.

[0139] 4. The controller serves as the core scheduling unit, implementing an energy management strategy of "photovoltaic priority, energy storage buffer." When photovoltaic energy is sufficient, the controller prioritizes directly supplying electricity to the power output unit, while simultaneously sending excess electricity to the energy storage unit through the energy management unit. When photovoltaic energy is insufficient, the controller automatically switches to energy storage power supply to ensure continuous and stable power output. The controller not only implements energy path control but also interacts with the edge computing unit and issues relevant control commands to the execution unit.

[0140] In practical implementation, when the controller executes the energy management strategy of "photovoltaic priority, energy storage buffer", the following judgment principles and control schemes can be adopted:

[0141] 1) Real-time monitoring and data acquisition

[0142] The controller collects the output voltage U of the photovoltaic module in real time through the sensing module. pv Current I pv The voltage U of the energy storage unit (battery) bat Current I bat The system also records the operating voltage and current requirements of the load side of the cathodic protection power supply unit. Simultaneously, it records the operating status parameters of each power electronic converter in the energy management unit.

[0143] 2) Judgment principles

[0144] When the photovoltaic output power P pv =U pv ×I pvThe power P required by the cathode protection power supply load is higher than that of the cathode protection power supply. load And the energy storage unit voltage U bat The upper limit threshold U was not reached. bat,max At that time, the "direct photovoltaic supply + surplus energy charging and storage" approach will be implemented;

[0145] When photovoltaic power is insufficient to meet P load , and U bat >U bat,min If so, then "energy storage compensation power supply" will be implemented;

[0146] When both photovoltaic power and energy storage capacity are insufficient, the controller issues a low power alarm signal and provides adjustment strategies through the edge computing unit to prioritize ensuring that the output potential of the cathodic protection power supply device is not lower than the set protection lower limit potential, thereby maintaining the minimum protection requirements of the cathodic protection circuit.

[0147] 3) Control scheme

[0148] The photovoltaic section can use control algorithms such as maximum power point tracking (MPPT) to dynamically adjust the operating point, so that the photovoltaic modules are kept in the optimal power output state;

[0149] The charging and discharging process can use voltage / current regulation algorithms (such as PI controller combined with PWM duty cycle regulation) to control the DC / DC step-up and step-down circuit, ensuring a smooth transition of voltage and current;

[0150] When U is detected pv >U load At this time, the controller prioritizes driving the photovoltaic-load path to conduct, and at the same time activates the energy management unit to import the remaining power into the energy storage unit;

[0151] When U is detected pv load In this case, the photovoltaic-load direct supply path will be automatically cut off, the energy storage unit-load path will be turned on, and the current will be kept continuous to avoid the cathodic protection power supply device being affected by potential changes.

[0152] 4) Data interaction and execution

[0153] The controller transmits real-time operating parameters such as voltage, current, and power to the edge computing unit. The edge computing unit then evaluates energy utilization efficiency and predicts trends based on the real-time operating parameters and historical operating records, and feeds optimization instructions back to the controller to achieve dynamic path switching and energy allocation. Finally, the controller sends PWM signals to control the power devices to turn on / off, completing the switching and scheduling of energy flow.

[0154] ​The judgment principles and control schemes described above in this invention work together with the MPPT function built into the photovoltaic module, the switching / conversion function of the battery management system (BMS) and the energy management unit to achieve optimal energy allocation for the entire system.

[0155] 5. The edge computing unit may include a multi-source data acquisition module and a self-identification pre-adjustment module. The multi-source data acquisition module collects protection potential, output current, output voltage, environmental parameters, and pipeline interference signals in real time; the self-identification pre-adjustment module performs local fusion calculations based on the acquired data, predicts potential deviation trends in advance, and generates adjustment commands to be sent to the controller or execution unit. This unit has independent local computing capabilities, and can still perform closed-loop calculations and output parameter corrections even in the event of a network interruption, avoiding complete dependence on cloud computing and improving the real-time performance and reliability of the device.

[0156] As an optional implementation of this invention, the edge computing unit performs local fusion calculations based on the collected data in the following manner to predict the potential deviation trend in advance and generate adjustment instructions:

[0157] Set the target protection interval [E] min E max Safety upper limit, safety lower limit, power and thermal constraints;

[0158] The collected data was subjected to first-order exponential smoothing and median filtering for noise reduction.

[0159] off (k)=λ off (k-1)+(1-λ) E off (k), λ∈[0.7,0.95]

[0160] in, off (k) represents the smoothed power-off potential of the kth sample, which is the denoised estimate obtained by first-order exponential smoothing filtering. λ is the smoothing coefficient (forgetting factor). When λ is large, the potential curve is more stable. When λ is small, the system responds more quickly to new sampling points.

[0161] Calculate instantaneous deviation:

[0162] e(k) = E set- off (k)

[0163] Where e(k) is the potential deviation of the k-th sampling (the difference between the set value and the measured power-off potential), E set Set the protection potential for the system.

[0164] Calculate the rate of change of deviation:

[0165] Δe(k) = e(k) - e(k-1)

[0166] Calculate trend strength:

[0167] Δe EWMA (k)=β Δe(k)+(1-β) Δe EWMA (k-1), β∈[0.3,0.7]

[0168] Where, Δe EWMA (k) represents the rate of change of the exponentially weighted moving average deviation of the k-th sample, and β is the smoothing weight factor.

[0169] The PID algorithm is used for online adjustment of control parameters:

[0170] u(k)=K p [e(k)-e(k-1)]+K i e(k)+K d [e(k)-2e(k-1)+e(k-2)]

[0171] Where u(k) is the controller output, the equivalent control input for directly driving the DC / DC or cathodic protection power supply, e(k) is the potential deviation of the k-th sampling (the difference between the set value and the measured power-off potential), and K p K is the proportionality coefficient. i K is the integral coefficient. d These are the differential coefficients;

[0172] By adding a trend feedforward and short-term prediction term before the PID output, the potential deviation value ê(k+1) of the (k+1)th prediction is obtained.

[0173] Calculate the pre-adjustment feedforward:

[0174] u ff (k) = αê(k+1), α > 0

[0175] Among them, u ff (k) is the trend feedforward control quantity, calculated from ê(k+1), used to compensate for potential change trends in advance, and α is the feedforward gain, used to adjust the weight of the feedforward component in the total control signal.

[0176] Perform synthesis control:

[0177] u pre (k)=u ff (k)+u(k)

[0178] Among them, u pre(k) is the pre-adjustment synthetic control signal, derived from the trend feedforward quantity u. ff The sum of (k) and the PID feedback quantity u(k) is used as the final control input to drive the output of the power output device.

[0179] For u pre (k) Apply soft / hard limiting and slope constraints:

[0180] u min ≤u pre (k)≤u max ,∣u pre (k)-u pre (k-1)∣≤r max

[0181] Among them, u min u is the lower limit value of the output control signal. max r is the upper limit of the output control signal. max This represents the upper limit of the output rate of change.

[0182] As an optional implementation of this invention, the edge computing unit uses a PID algorithm to adjust the control parameters online in the following manner:

[0183] The discrete form and parameter tuning rules include: using the potential deviation value e and the rate of change of potential deviation Δe as inputs, adjusting the proportional coefficient K online. p Integral coefficient K i and differential coefficient K d ;

[0184] When |e|>0.1V, increase the proportionality coefficient K. p Decrease the differential coefficient K d To quickly eliminate bias, and then reduce the integral coefficient K after stabilization. i Suppress overshoot;

[0185] When |e| < 0.1 V, decrease the proportionality coefficient K. p Differential coefficient K d Improve steady-state accuracy.

[0186] As an optional implementation of this invention, the edge computing unit adds a trend feedforward and short-term prediction term before the PID output in the following manner:

[0187] Using moving linear prediction:

[0188] ê(k+1) =ae(k)+(1-a)e(k-1), a∈[0.6,0.9]

[0189] Where ê(k+1) is the predicted potential deviation value at time k+1, and a is the moving linear prediction weight coefficient;

[0190] Or with Δe EWMA Take a forward look:

[0191] ê(k+1)=e(k)+Δe EWMA (k).

[0192] In specific implementation, the edge computing unit provided in this embodiment of the invention has the following characteristics:

[0193] 1) Variables and Signals

[0194] Data collected: Protection potential E on E off AC voltage V AC Output current I o Output voltage U o Environmental parameters (T, soil moisture content, soil resistivity ρ, pH); external disturbance indicators (track voltage / AC / geomagnetic index, etc.).

[0195] Benchmarks and Constraints: Target Protection Zone [E] min E max (e.g., −1.20V≤ E) off ≤−0.85V); safety upper / lower limits, power and thermal constraints.

[0196] Preprocessing (first-order exponential smoothing and median filtering for noise reduction):

[0197] off (k)=λ off (k-1)+(1-λ) E off (k), λ∈[0.7,0.95]

[0198] 2) Deviation and Trend Quantity

[0199] Instantaneous deviation:

[0200] e(k) = E set- off (k)

[0201] Rate of change of deviation (discrete derivative):

[0202] Δe(k) = e(k) - e(k-1)

[0203] Trend strength (EWMA derivative):

[0204] Δe EWMA(k)=β Δe(k)+(1-β) Δe EWMA (k-1), β∈[0.3,0.7]

[0205] 3) Fuzzy Adaptive PID

[0206] In this embodiment, the self-identification pre-adjustment module selects the PID algorithm to adjust the control parameters online.

[0207] Discrete PID control law (power supply side given voltage or duty cycle equivalent voltage):

[0208] u(k)=K p  [e(k)-e(k-1)]+K i e(k)+K d [e(k)-2e(k-1)+e(k-2)]

[0209] Where: u(k) is the equivalent control input (duty cycle / given voltage) for directly driving a DC / DC or cathodic protection power supply. This discrete form and parameter tuning rules include: using e and Δe as inputs, adjusting K online. p , K i , K d When |e|>0.1V, increase K. p Decrease K d To quickly eliminate bias and stabilize the system, K should be reduced. i Suppress overshoot; decrease K when |e| < 0.1 V. p , K d Improve steady-state accuracy.

[0210] 4) Pre-adjustment feedforward

[0211] To reflect the principle of "preemptive judgment and correction," trend feedforward and short-term prediction terms are added before the PID output:

[0212] a) Sliding linear prediction (lightweight implementation at the edges)

[0213] ê(k+1)=ae(k)+(1-a)e(k-1), a∈[0.6,0.9]

[0214] Or with Δe EWMA Make a forward look: ê(k+1) = e(k) + Δe EWMA (k).

[0215] b) Pre-adjusted feedforward

[0216] u ff (k) = αê(k+1), α > 0

[0217] 5) Synthesis control

[0218] u pre (k)=u ff (k)+u(k)

[0219] And for u pre (k) Apply soft / hard limiting and slope control (to prevent overshoot and electrochemical shock):

[0220] u min ≤u pre (k)≤u max ,∣u pre (k)-u pre (k-1)∣≤r max

[0221] As an optional embodiment of the present invention, the controller is further configured to appropriately increase the differential coefficient K when a significant fundamental frequency component or daily periodic energy peak is detected in the potential signal. d Reduce the integral coefficient K i And reduce the feedforward gain α.

[0222] As an optional embodiment of the present invention, the controller is further configured to temporarily increase the proportional coefficient K when the rate of change of the potential deviation |Δe| exceeds a preset threshold. p And increase the feedforward gain α.

[0223] In specific implementation, the present invention also has a pre-adjustment strategy driven by interference recognition:

[0224] Periodic interference scenarios (such as AC interference or tidal interference): When a significant fundamental frequency component or daily periodic energy peak is detected in the potential signal, the module enters a periodic external disturbance mode. At this time, the controller will appropriately increase the differential coefficient K. d Reduce the integral coefficient K i Furthermore, the feedforward gain α is reduced to enhance the ability to suppress periodic disturbances and maintain steady-state accuracy.

[0225] In step-type disturbance scenarios (such as drainage device switching, bridging connection, or periodic power-on / power-off operations): when the rate of change of potential deviation |Δe| exceeds a preset threshold, the module enters step-type disturbance mode. At this time, the controller will temporarily increase the proportional coefficient K. p It also increases the feedforward gain α to quickly offset sudden changes; after the interference decays or terminates, the parameters automatically return to the normal tuning value to ensure system stability.

[0226] 6. The execution unit may include an interference suppression module and an intelligent graded early warning module. The interference suppression module combines hardware filtering and dynamic compensation methods to effectively counteract the impact of DC interference from rail transit, high-voltage DC grounding electrodes, and ground current interference on the cathodic protection power supply device. The intelligent graded early warning module can automatically classify the risk level according to the intensity and duration of the interference and trigger audible and visual alarms or local prompts, thereby achieving rapid on-site response and maintenance. The execution unit receives adjustment commands from the edge computing unit and directly applies the execution results to the power output device to adjust the output potential of this power supply device.

[0227] As an optional embodiment of the present invention, the execution unit receives the adjustment command in the following manner, performs hardware filtering and dynamic compensation on the output control signal, and applies the compensated control quantity to the power output device:

[0228] The interference is decomposed into interference components in different frequency bands, such as quasi-DC / slow drift, low-frequency oscillation, power frequency and harmonics, and the corresponding dynamic compensation amount is output for each.

[0229] The dynamic compensation is superimposed on the output signal of the power supply control circuit and applied directly to the power output device; or, the dynamic compensation is injected into the equivalent setting port or PWM duty cycle control port of the power output device.

[0230] As an optional implementation of this invention, the execution unit outputs the DC offset compensation amount in the following manner:

[0231] b(k+1)=b(k)+w(k), w~N(0,Q)

[0232] E dc (k)=E true (k)+b(k)+v(k), v~N(0,R)

[0233] Kalman filtering or exponential smoothing methods are used to address slow drift. (k) Estimate the offset to obtain the estimated value and generate the corresponding DC compensation amount:

[0234] u dc (k)=K dc (k), 0 <K dc ≤1

[0235] Where b(k+1) is the slow drift or zero bias estimate term calculated in the (k+1)th time, w(k) is the process noise term, and E dc (k) represents the DC potential measurement value from the kth sampling, E true (k) represents the true DC potential, v(k) represents the measurement noise term, and u dc(k) DC compensation control quantity, K dc This is the DC compensation gain coefficient.

[0236] As an optional implementation of this invention, the execution unit outputs the AC interference compensation amount in the following manner:

[0237] The disturbance is approximated as a superposition of sinusoidal bases:

[0238]

[0239] ω m Tracked by PLL / FLL; LMS updates minimize residuals:

[0240] ε(k)=E ac (k)- ,

[0241] Calculate the compensation amount:

[0242]

[0243] in, E is the estimated value of the AC interference signal from the k-th sample; ac (k) represents the measured AC potential signal from the kth sampling. Let be the orthogonal basis vector of the m-th interference component; The angular frequency of the m-th interference component is determined by PLL / FLL tracking; T s The sampling period; w m (k) : These are the weight coefficient vectors of the m-th component and their estimated values, respectively; The step size coefficient of the LMS algorithm controls the weight update speed and convergence stability; This is the residual signal, which is the difference between the actual measured signal and the estimated signal; The amount of compensation for AC interference is the negative value of the estimated interference to achieve cancellation.

[0244] As an optional implementation of this invention, the execution unit outputs the periodic disturbance compensation amount in the following manner:

[0245] Periodic disturbance compensation is adopted:

[0246] u rep (k)=u rep (kN)+Q(z) e(kN),

[0247] Where Q(z) is a low-pass stabilization filter, urep (k) is the periodic disturbance compensation amount, e(k) is the potential deviation, and T period For the interference period, T s The sampling period.

[0248] As an optional implementation of this invention, the execution unit outputs the feedforward compensation amount in the following manner:

[0249] Reference feedforward compensation is adopted:

[0250] u ff (k) = G ff (z) d ref (k)

[0251] Among them, G ff (z) is the feedforward transfer function determined by calibration or identification; d ref (k) is the reference disturbance signal; u ff (k) represents the reference feedforward compensation amount.

[0252] As an optional implementation of this invention, the execution unit is further configured to synthesize compensation and limit amplitude / slope:

[0253] u raw (k)=u dc (k)+u ac (k)+u rep (k)+u ff (k)

[0254] u comp (k)=clip (u raw (k), u min ,u max ),∣u comp (k)-u comp (k-1)∣≤r max

[0255] Among them, u raw (k) represents the original synthesized compensation signal without amplitude limiting processing; u comp (k) is the dynamically compensated synthesized signal output by the execution unit; u dc (k) represents the DC offset compensation amount; u ac (k) represents the AC interference compensation amount; u rep (k) represents the periodic disturbance compensation amount; u ff (k) represents the trend feedforward compensation amount; u min u max These are the lower and upper limits of the output limiting, respectively; r max This represents the upper limit of the output rate of change.

[0256] When predicting off If the gain is about to exceed the limit, the adaptive gain is temporarily frozen, and a minimum compensation is applied while the recovery is gradually released.

[0257] In specific implementation, the execution unit provided in this embodiment of the invention has the following characteristics:

[0258] 1) Principles and models of execution units (discrete time domain)

[0259] Sampling period T s At time k, the target potential E set Measured potential E meas (k), error e(k) = E set- E meas (k) decomposes the interference into: quasi-DC / slow drift d dc (k) (refers to slow potential shifts below 0.005 Hz), low-frequency oscillations d lf (k), power frequency and harmonics d ac (k) (E obtained by digital filtering after hardware RC / LC anti-aliasing) meas (k)=E true (k)+d dc (k)+d lf (k)+d ac (k)+n(k)

[0260] The dynamic compensation amount output by the execution unit is u comp (k) is superimposed on the output signal of the constant potential control loop and directly applied to the output terminal of the cathodic protection power supply device; in another implementation, the compensation amount can also be injected into the equivalent setting port or PWM duty cycle control port of the cathodic protection power supply device, thereby achieving fast interference suppression without changing the structure of the main control loop.

[0261] 2) Channel-specific compensation algorithm

[0262] a) Quasi-DC / Slow Drift Compensation

[0263] b(k+1)=b(k)+w(k), w~N(0,Q)

[0264] E dc (k)=E true (k)+b(k)+v(k), v~ N(0,R)

[0265] Estimation using Kalman or exponential smoothing (k) generates compensation

[0266] u dc (k)=K dc (k), 0 <K dc ≤1

[0267] b) Power frequency / harmonic compensation

[0268] The disturbance is approximated as a superposition of sinusoidal bases:

[0269]

[0270] ω m Tracked by PLL / FLL; LMS updates minimize residuals:

[0271] ε(k)=E ac (k)- ,

[0272] Compensation amount:

[0273]

[0274] c) Periodic disturbance compensation (applicable to disturbances characterized by tidal / diurnal variations)

[0275] u rep (k)=u rep (kN)+Q(z) e(kN),

[0276] Q(z) is a low-pass stabilization filter.

[0277] d) Reference feedforward compensation (optional, when there is track / grounding electrode / pipe ground AC measurement)

[0278] u ff (k) = G ff (z) d ref (k)

[0279] Among them G ff (z) Determined by calibration / identification (amplitude matching).

[0280] 3) Synthesis and Constraint Mechanisms

[0281] Synthetic compensation and amplitude / slope limiting to avoid overcompensation:

[0282] u raw (k)=u dc (k)+u ac (k)+u rep (k)+u ff (k)

[0283] u comp(k)=clip (u raw (k), u min ,u max ),∣u comp (k)-u comp (k-1)∣≤r max

[0284] Injected in parallel with the main circuit of the power supply unit, shortening the response link; when predicting off If the gain is about to exceed the limit, temporarily freeze the adaptive gain (e.g., set μ → 0, limit K). dc (Change quantity), switch to a safety margin compensation and slow recovery to prevent overshoot and oscillation. Where clip(x, x) min x max ) represents the amplitude limiting function; r max The maximum allowable rate of output change.

[0285] 7. The output of this power supply can operate according to a preset mode to maintain the potential of the interface between the protected pipe or metal structure and the electrolyte within a set range, thereby effectively suppressing electrochemical corrosion. The output is connected to the controller and the execution unit, ensuring the synergistic effect of steady-state power supply and dynamic adjustment.

[0286] The cathode protection power supply device based on photovoltaic power supply and edge computing control provided in this invention adopts an energy management structure of "photovoltaic priority and battery buffer". It achieves direct photovoltaic load driving, surplus power storage, and energy storage supplementation when photovoltaic power is insufficient through bidirectional DC / DC and automatic switching control, thus forming a closed-loop energy management path with direct photovoltaic load supply as the main component and bidirectional interaction between photovoltaic and energy storage as a supplement. This not only avoids dual energy conversion but also reduces unnecessary battery charging and discharging, resulting in two effects: firstly, improved overall energy efficiency and reduced heat generation and loss; secondly, reduced battery cycle depth and frequency, extending battery life, while maintaining the stability of the power output potential.

[0287] Furthermore, the device integrates an edge computing unit, with multi-source data acquisition interfaces and a self-identification pre-adjustment module configured in parallel, enabling a closed-loop processing flow for acquisition, fusion, computation, and control locally. Combined with hardware filtering and dynamic compensation interference suppression mechanisms, and the coordinated execution of the intelligent hierarchical early warning module, even during network interruptions or in complex electromagnetic environments such as rail AC / DC, high-voltage DC grounding electrodes, and stray ground current interference, the system can still complete data processing and control decisions locally, correct the output potential, and simultaneously generate risk level alerts, ensuring timely and continuous stability of the protection status. As a result, the device significantly improves on-site response speed, anti-interference stability, offline availability, and adaptability to complex operating conditions, meeting the requirements for long-term reliable operation in environments without mains power or with strong interference.

[0288] Therefore, the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in this embodiment of the invention constructs an energy management structure that prioritizes photovoltaic power and buffers battery power. When the photovoltaic output meets the power consumption requirements of the cathodic protection power supply, it directly supplies power from the photovoltaic system and stores the remaining energy in the energy storage unit. When the photovoltaic output is insufficient, it automatically switches to power supply from the energy storage unit, reducing dual energy conversion losses, lowering the battery charging and discharging frequency, extending the energy storage life, and improving the stability of the output potential of the cathodic protection power supply device. At the same time, the device is equipped with an edge computing unit, which integrates multi-source data acquisition interfaces and a self-identification pre-adjustment module in parallel to achieve local fusion, rapid calculation, and parameter correction, and can operate stably even when the network is down. An integrated execution module for interference suppression and intelligent hierarchical early warning is set up, which uses hardware filtering and dynamic compensation, and completes risk level determination and hierarchical alarm locally, improving anti-interference stability and risk visualization.

[0289] Figure 2 This document illustrates a flowchart of a control method for a cathodic protection power supply device based on photovoltaic power supply and edge computing, provided in an embodiment of the present invention. This method is applied to the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in the above embodiment. The following only describes the flow of the method; for other matters not covered herein, please refer to the relevant descriptions of the cathodic protection power supply device based on photovoltaic power supply and edge computing control provided in the above embodiment. Further details will not be elaborated here. See [link to documentation]. Figure 2 The control method for a cathode protection power supply device based on photovoltaic power supply and edge computing provided in this embodiment of the invention includes:

[0290] S1 collects photovoltaic, energy storage, potential, current, voltage and interference signals;

[0291] S2, implements a photovoltaic priority energy management strategy;

[0292] S3 performs local denoising, feature extraction, and trend prediction on potential and interference signals;

[0293] S4 generates adjustment commands based on deviation, rate of change, and trend.

[0294] S5 performs DC compensation, AC compensation, periodic disturbance compensation, and feedforward compensation;

[0295] S6, synthesize the compensation amount and apply amplitude and slope limiting to the output;

[0296] S7 determines the risk level and issues a warning based on the intensity and duration of interference.

[0297] Therefore, the control method for a cathodic protection power supply device based on photovoltaic power supply and edge computing provided in this embodiment of the invention constructs an energy management structure that prioritizes photovoltaic power and buffers battery power. When the photovoltaic output meets the power consumption requirements of the cathodic protection power supply, the photovoltaic power supply is directly supplied, and the remaining electrical energy is stored in the energy storage unit. When the photovoltaic output is insufficient, the power supply is automatically switched to the energy storage unit, reducing the dual energy conversion loss, lowering the battery charging and discharging frequency, extending the energy storage life, and improving the stability of the output potential of the cathodic protection power supply device. At the same time, the device is equipped with an edge computing unit, which integrates multi-source data acquisition interfaces and a self-identification pre-adjustment module in parallel to achieve local fusion, fast calculation, and parameter correction, and can operate stably even when the network is disconnected. An integrated execution module for interference suppression and intelligent hierarchical early warning is set up, which uses hardware filtering and dynamic compensation, and completes risk level determination and hierarchical alarm locally, improving anti-interference stability and risk visualization.

[0298] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0299] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A cathodic protection power supply device based on photovoltaic power supply and edge computing, characterized in that, include: Photovoltaic modules, controllers, energy management units, edge computing units, execution units, energy storage units, and power output devices; among which: The photovoltaic module is electrically connected to the energy management unit and is used to convert solar energy into direct current (DC) power and provide the DC power to the energy management unit. An energy management unit, electrically connected to the energy storage unit and the controller, is used to interact with the energy storage unit, charge and discharge the energy storage unit, and distribute energy under the scheduling of the controller; The controller is used to collect the output voltage and current of the photovoltaic module, the voltage and current of the energy storage unit, and the load demand of the cathodic protection power supply. When the photovoltaic output power is higher than the load demand of the cathodic protection power supply and the voltage of the energy storage unit has not reached the upper limit threshold, the controller controls the energy management unit to use the DC power converted by the photovoltaic module to supply power and charge the energy storage unit. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is greater than the lower limit threshold, the controller controls the energy management unit to disconnect the DC power supply path converted by the photovoltaic module and use the energy storage unit to supply power. When the photovoltaic output power does not meet the load demand of the cathodic protection power supply and the voltage of the energy storage unit is less than or equal to the lower limit threshold, a low battery alarm signal is generated. The controller also receives adjustment instructions from the edge computing unit to maintain the output potential of the cathodic protection device at no lower than the set protection lower limit potential. The edge computing unit is used to collect protection potential, output current, output voltage, environmental parameters and pipeline interference signals in real time, perform local fusion calculations based on the collected data, predict the potential deviation trend in advance, and generate the adjustment command to be sent to the controller or the execution unit. The execution unit is used to receive the adjustment command, perform hardware filtering and dynamic compensation on the output control signal, and apply the compensated control quantity to the power output device; at the same time, it classifies the risk level according to the intensity and duration of the interference and triggers an alarm. The power output device is used to provide the power output signal regulated by the execution unit to the external cathodic protection circuit; The edge computing unit performs local fusion calculations based on the collected data in the following manner to predict the potential deviation trend in advance and generate adjustment commands: Set the target protection interval [E] min E max Safety upper limit, safety lower limit, power and thermal constraints; The collected data was subjected to first-order exponential smoothing and median filtering for noise reduction. off (k)=λ off (k-1)+(1-λ) E off (k), λ∈[0.7,0.95] in, off (k) is the smoothed power-off potential of the kth sample, which is the denoised estimate obtained by first-order exponential smoothing filter, and λ is the smoothing coefficient. Calculate instantaneous deviation: e(k)=E set - off (k) Where e(k) is the potential deviation of the k-th sampling, E set Set the protection potential for the system; Calculate the rate of change of deviation: Δe(k) = e(k) - e(k-1) Calculate trend strength: Δe EWMA (k)=β Δe(k)+(1-β) Δe EWMA (k-1),β∈[0.3,0.7] Where, Δe EWMA (k) represents the rate of change of the exponentially weighted moving average deviation in the kth iteration, and β is the smoothing weight factor; The PID algorithm is used for online adjustment of control parameters: u(k)=K p [e(k)-e(k-1)]+K i e(k)+K d [e(k)-2e(k-1)+e(k-2)] Where u(k) is the controller output, the equivalent control input for directly driving the DC / DC or cathodic protection power supply, e(k) is the potential deviation of the k-th sample, and K p K is the proportionality coefficient. i K is the integral coefficient. d These are the differential coefficients; By adding a trend feedforward and short-term prediction term before the PID output, we get ê(k+1); Calculate the pre-adjustment feedforward: u ff (k)=αê(k+1), α>0 Among them, u ff (k) is the trend feedforward control quantity, and α is the feedforward gain coefficient; Perform synthesis control: in pre (k)=u ff (k)+u(k) Among them, u pre (k) is the pre-adjusted synthetic control signal; For u pre (k) Apply soft / hard limiting and slope constraints: you min ≤u pre (k)≤u max ,∣u pre (k)-u pre (k-1)∣≤r max Among them, u min u is the lower limit value of the output control signal. max r is the upper limit of the output control signal. max This represents the upper limit of the output rate of change.

2. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 1, characterized in that, The edge computing unit uses a PID algorithm to adjust control parameters online in the following manner: The discrete form and parameter tuning rules include: using the potential deviation value e and the rate of change of potential deviation Δe as inputs, adjusting the proportional coefficient K online. p Integral coefficient K i and differential coefficient K d ; When |e|>0.1V, increase the proportionality coefficient K. p Decrease the differential coefficient K d To quickly eliminate bias, and then reduce the integral coefficient K after stabilization. i Suppress overshoot; When |e| < 0.1 V, decrease the proportionality coefficient K. p Differential coefficient K d Improve steady-state accuracy.

3. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 2, characterized in that, The edge computing unit adds trend feedforward and short-term prediction terms before the PID output in the following manner: Using moving linear prediction: ê(k+1) =ae(k)+(1-a)e(k-1), a∈[0.6,0.9] Where ê(k+1) is the predicted potential deviation value at time k+1, and a is the moving linear prediction weight coefficient; Or with Δe EWMA Take a forward look: ê(k+1)=e(k)+Δe EWMA (k)。 4. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 3, characterized in that, The controller is also configured to appropriately increase the differential coefficient K when a significant fundamental frequency component or daily periodic energy peak is detected in the potential signal. d Reduce the integral coefficient K i And reduce the feedforward gain α.

5. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 4, characterized in that, The controller is also configured to temporarily increase the proportional coefficient K when the rate of change of the potential deviation |Δe| exceeds a preset threshold. p And increase the feedforward gain α.

6. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 5, characterized in that, The execution unit receives the adjustment command in the following manner, performs hardware filtering and dynamic compensation on the output control signal, and applies the compensated control quantity to the power output device: The interference is decomposed into interference components of different frequency bands, including quasi-DC / slow drift, low-frequency oscillation, power frequency and harmonics, and the corresponding dynamic compensation amount is output for each. The output signal of the power supply control circuit is superimposed on the output signal of the power supply device and applied directly to the power output device; Alternatively, the dynamic compensation amount can be injected into the equivalent setting port or PWM duty cycle control port of the power output device.

7. The cathodic protection power supply device based on photovoltaic power supply and edge computing according to claim 6, characterized in that, The execution unit outputs the DC offset compensation amount in the following manner: b(k+1)=b(k)+w(k), w~N(0,Q) E dc (k)=E true (k)+b(k)+v(k) , v~ N(0,R) Kalman filtering or exponential smoothing methods are used to address slow drift. (k) Estimate the offset to obtain the offset estimate and generate the corresponding DC compensation amount: you dc (k)=K dc (k) , 0 <K dc ≤1 Where b(k+1) is the slow drift or zero bias estimate term calculated in the (k+1)th time, w(k) is the process noise term, and E dc (k) represents the DC potential measurement value from the kth sampling, E true (k) represents the true DC potential, v(k) represents the measurement noise term, and u dc (k) is the DC compensation control quantity, K dc This is the DC compensation gain coefficient.

8. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 6, characterized in that, The execution unit outputs the AC interference compensation amount in the following manner: The disturbance is approximated as a superposition of sinusoidal bases: ω m Tracked by PLL / FLL; LMS updates minimize residuals: e(k)=E ac (k)- , Calculate the compensation amount: in, E is the estimated value of the AC interference signal from the k-th sample; ac (k) represents the measured AC potential signal from the kth sampling. Let be the orthogonal basis vector of the m-th interference component; The angular frequency of the m-th interference component is determined by PLL / FLL tracking; T s The sampling period; w m (k) : These are the weight coefficient vectors of the m-th component and their estimated values, respectively; The step size coefficient of the LMS algorithm controls the weight update speed and convergence stability; This is the residual signal, which is the difference between the actual measured signal and the estimated signal; The amount of compensation for AC interference is the negative value of the estimated interference to achieve cancellation.

9. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 6, characterized in that, The execution unit outputs the periodic disturbance compensation amount in the following manner: Periodic disturbance compensation is adopted: Where Q(z) is a low-pass stabilization filter, u rep (k) is the periodic disturbance compensation amount, e(k) is the potential deviation, and T period For the interference period, T s The sampling period.

10. The cathode protection power supply device based on photovoltaic power supply and edge computing according to claim 6, characterized in that, The execution unit outputs the feedforward compensation amount in the following manner: Reference feedforward compensation is adopted: u ff (k) = G ff (z) d ref (k) Among them, G ff (z) is the feedforward transfer function determined by calibration or identification, d ref (k) is the reference disturbance signal, u ff (k) represents the reference feedforward compensation amount.

11. The cathode protection power supply device based on photovoltaic power supply and edge computing according to any one of claims 7-10, characterized in that, The execution unit is also used to synthesize compensation and limit amplitude / slope: in raw (k)=u dc (k)+u ac (k)+u rep (k)+u ff (k) you comp (k)=clip (u raw (k), you min ,you max ),∣u comp (k)-u comp (k-1)∣≤r max Among them, u raw (k) represents the original synthetic compensation signal without amplitude limiting processing; u comp (k) is the dynamic compensation synthesized signal output by the execution unit; u dc (k) represents the DC offset compensation amount; u ac (k) represents the AC interference compensation amount; u rep (k) represents the periodic disturbance compensation amount; u ff (k) represents the trend feedforward compensation amount; u min u max These are the lower and upper limits of the output limiting, respectively; r max This is the upper limit of the output rate of change; When predicting off If the gain is about to exceed the limit, the adaptive gain is temporarily frozen, and a minimum compensation is applied while the recovery is gradually released.

12. A control method for a cathodic protection power supply device based on photovoltaic power supply and edge computing, characterized in that, The cathode protection power supply device based on photovoltaic power supply and edge computing as described in any one of claims 1 to 11 includes: Collects photovoltaic, energy storage, potential, current, voltage, and interference signals; Implement a photovoltaic-priority energy management strategy; Local denoising, feature extraction, and trend prediction are performed on potential and interference signals. Adjustment instructions are generated based on deviation, rate of change, and trend. Performs DC compensation, AC compensation, periodic disturbance compensation, and feedforward compensation; Synthesize the compensation amount and apply amplitude and slope limiting to the output; The risk level is determined and an early warning is issued based on the intensity and duration of the interference.

Images