A cooperative energy supply cold region railway subgrade frost damage prevention system and operation method

Abstract

A synergistic energy supply system for preventing frost damage to railway subgrades in cold regions and its operation method are disclosed. The system includes a photovoltaic-photothermal composite array unit, a heat transfer and circulation unit, an electrical energy storage and power supply unit, and an intelligent control unit. The photovoltaic-photothermal composite array simultaneously outputs DC electrical energy and thermal energy in the same plane. The thermal energy is circulated through a circulation pump to drive antifreeze to the heating rods within the subgrade for release. The electrical energy is stored in a battery and can power the nano-heating core integrated within the heating rods. The intelligent control unit automatically switches between photothermal circulation heating mode and resistance heating mode based on temperature thresholds and hysteresis logic, achieving all-weather adaptive active heating. This invention achieves synergistic solar photovoltaic-photothermal energy supply, resulting in high energy efficiency, continuous and reliable heating, and eliminating the need for an external power grid, making it suitable for preventing frost damage to railway subgrades in cold regions.

Description

Technical Field

[0001] This invention relates to the field of disaster prevention and mitigation technology for railway engineering in cold regions, specifically to a collaborative energy supply system for preventing frost damage to railway subgrade in cold regions and its operation method. Background Technology

[0002] Frost damage to railway subgrades in cold regions, particularly frost heave caused by the freezing of soil moisture, is a core threat to the safety and smoothness of railway operations. To prevent and control this problem, existing technologies have evolved from passive insulation and structural reinforcement to third-generation technologies centered on active heating. The basic function of these technologies is to actively transfer heat to the subgrade bed to counteract the effects of the cold external environment, maintain the soil temperature above the freezing point, thereby inhibiting frost heave and ensuring the construction quality and operational safety of railways in cold regions (such as the Sichuan-Tibet Railway and the Qinghai-Tibet Railway).

[0003] Currently, those skilled in the art have proposed various active heating technology solutions. Early patents, such as the solar thermal collector device for roadbed engineering in seasonally frozen soil areas (CN201820341454.9) and the internal circulation solar heating device for roadbeds (CN201811518147.4), mainly employ independent solar thermal systems. These systems absorb heat through solar collectors, and the heat medium is driven by circulation pipelines and pumps to circulate between the collectors and heat dissipation pipes buried in the roadbed, directly delivering the heat to the interior of the roadbed. More recent technologies, such as the solar photovoltaic power-to-heat roadbed anti-frost damage device for seasonally frozen areas (CN119332556A), use photovoltaic power generation to drive electric heating pipes to heat the roadbed, and integrate components such as battery energy storage and water flow cleaning of the photovoltaic panels. In addition, the railway subgrade frost heave prevention and control system and method adapted to severe cold climates (CN120401292A) uses a ground source heat pump as its core. It extracts low-grade geothermal energy from the stable foundation layer through a compressor, upgrades the grade, and releases heat to the frost heave-prone layer of the subgrade. Multiple auxiliary branches are set up to solve the compressor's start-up and operation problems in severe cold environments. There is also an anti-frost heave subgrade system and subgrade heating method (CN115467206A), which combines geothermal energy devices to collect underground heat energy and uses solar photovoltaic panels to power the heat pump compressor and other electrical equipment, achieving partial energy self-sufficiency.

[0004] However, existing technologies still suffer from the following core shortcomings. First, the overall energy utilization efficiency is low, failing to maximize the value of solar energy: pure solar thermal systems only utilize the sun's thermal energy, neglecting its solar energy component; pure photovoltaic systems only convert solar energy into electrical energy and then into thermal energy, resulting in a long and inefficient energy conversion path, and failing to effectively utilize the solar thermal energy component, leading to limited energy output per unit area of ​​solar energy collection devices, and failing to achieve the synergistic utilization and cascade recovery of solar energy through "photovoltaics and solar thermal". Second, the all-weather active heating guarantee capability is insufficient: pure solar thermal systems are essentially ineffective at night and on cloudy days, relying only on limited residual heat, resulting in severely insufficient heating continuity; pure photovoltaic systems relying on batteries are completely limited in heating capacity and lifespan at night or during consecutive cloudy days, posing a risk of insufficient power supply; ground source heat pump systems face challenges in operating efficiency and stability under extreme cold conditions and rely on grid power. Existing technologies generally lack a low-cost, high-efficiency all-weather energy supply solution, making it difficult to achieve the core goal of "stable heating throughout the cold season". Third, the systems are functionally limited and highly dependent on external systems: most systems lack power generation capabilities, and their control units, circulating pumps, and other electrical equipment must rely on the external power grid, making them difficult to apply in remote, cold regions without a power grid. Fourth, the balance between system complexity and economy is poor, and the level of intelligent control needs improvement: some systems integrate multiple complex subsystems to enhance functionality (such as water flow cleaning power generation systems and multi-branch heat pump systems), leading to high costs, complex maintenance, and increased reliability risks in extremely cold environments; at the same time, their control strategies are either relatively simple (such as based on a single ambient temperature) or focus on optimizing the operation of the equipment itself, lacking the intelligent decision-making ability to adaptively and precisely supply heat on demand based on the actual thermal state of the roadbed and the system's own energy state. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention discloses a collaborative energy supply system and its operation method for preventing frost damage to railway subgrades in cold regions. The technical solution is as follows:

[0006] A collaborative energy supply system for preventing frost damage to railway subgrade in cold regions, characterized in that it includes:

[0007] A photovoltaic-photothermal composite array unit is installed on the slope or open space along a railway line to simultaneously output DC power and heat energy. The unit includes a photovoltaic panel and a photothermal liquid storage unit integrated below the photovoltaic panel. The photothermal liquid storage unit is filled with antifreeze to absorb solar radiation energy not utilized by the photovoltaic panel and waste heat generated when the photovoltaic panel is working, thereby realizing photothermal conversion and cooling the photovoltaic panel.

[0008] The heat transfer and circulation unit includes a heating rod, a circulation pump, and an upward and downward conduit connecting the photothermal liquid storage unit and the heating rod, all buried in the subgrade bed. The heating rod has a hollow pipe inside for the antifreeze to flow through, and its outer wall is thermally conductive. The circulation pump is used to drive the antifreeze to circulate in a closed loop formed by the photothermal liquid storage unit, the upward conduit, the heating rod, and the downward conduit, so as to transfer heat to the subgrade soil.

[0009] The energy storage and power supply unit includes a photovoltaic inverter and controller, a battery bank, and cables laid to the heating rod; the photovoltaic inverter and controller is used to stabilize and rectify the DC power generated by the photovoltaic panel and manage the charging and discharging of the battery bank; the battery bank is used to store excess energy generated by photovoltaic power generation.

[0010] The intelligent control unit includes a temperature probe installed in the photothermal liquid storage unit and a remote intelligent control module; the remote intelligent control module intelligently controls the start and stop of the circulation pump and the heating wire according to the temperature probe signal and the preset program, so as to realize the automatic switching of the working mode.

[0011] The heating rod has an integrated heating wire inside or outside. The heating wire is electrically connected to the power supply unit via a cable and is used to convert electrical energy into heat energy. The heating wire is a nano-heating core with low power consumption and high heat dissipation.

[0012] This invention also discloses an operation method for preventing frost damage to railway subgrade in cold regions based on the above system, characterized by the following steps:

[0013] System initialization and self-test steps: After the system is powered on, it performs temperature sensor validity detection, battery power detection, equipment fault status detection and communication initialization. After all of them pass, it enters the main control loop.

[0014] Main control loop steps: Perform temperature reading, mode switching, fault detection, energy efficiency optimization and status reporting periodically at fixed time intervals of 30 seconds;

[0015] Mode switching steps: Real-time monitoring of antifreeze temperature Based on the battery's state of charge (SOC), a dynamic hysteresis algorithm is used to calculate the switching threshold and the dynamic hysteresis amount. Based on the rate of temperature change Adjustment, unit ℃ / s: when hour, ;otherwise ;when When the time comes, start the circulation pump and turn off the heating element; when and When the time comes, turn off the circulation pump and start the heating element; when At that time, shut down all heating equipment;

[0016] Adaptive prediction and pre-switching steps: If it is predicted that the solar irradiance will drop below 50 W / m² within the next hour and the current system is in photothermal cycle mode, the heating element will be activated in advance for preheating to maintain the antifreeze temperature above 25°C; if it is predicted that there will be more than 48 consecutive cloudy days in the future, the system will be forced to switch to energy-saving mode, lowering the heating element activation threshold to 25°C and increasing the battery protection capacity to 30%.

[0017] Fault self-diagnosis and fault tolerance steps: Detect abnormal states of temperature sensor, circulation pump, and heating wire in each control cycle; when a temperature sensor fault is detected, automatically switch to a virtual sensor model based on historical temperature and irradiance to estimate antifreeze temperature; when a circulation pump fault is detected, automatically switch to pure resistance heating mode.

[0018] Energy efficiency optimization and thermal balance linkage steps: Calculate the instantaneous comprehensive energy efficiency ratio of the system every 10 minutes. and the heat required for the roadbed Construct a multi-objective optimization function:

[0019]

[0020] in, These are the weighting coefficients. This is the theoretical maximum energy efficiency ratio. For actual heat supply, This refers to the battery discharge rate.

[0021] The optimal combination of circulating pump speed and heating wire duty cycle is determined through online optimization.

[0022] On-demand precise heating steps: In resistance heating mode, based on the optimal duty cycle The heating element is controlled by a pulse width modulation signal, and the actual heating power is... and make it match ;

[0023] Battery health management steps: When the battery health level... When the battery temperature is below 0°C, the heating element will be activated to intermittently heat the battery compartment to above 5°C.

[0024] Status reporting and remote optimization steps: A status record is generated every 30 seconds and uploaded to the cloud monitoring platform via wireless communication. The platform uses a time series prediction model to continuously optimize the control parameters for the next day based on the uploaded data and then distributes the results.

[0025] Beneficial effects

[0026] 1. Energy utilization efficiency has been significantly improved.

[0027] By integrating photovoltaic panels and photothermal liquid storage units into a "photovoltaic-photothermal composite array," photoelectric conversion and photothermal conversion are simultaneously achieved in the same space, effectively recovering the waste heat generated during photovoltaic panel operation and realizing the cascade utilization of the entire solar spectrum. Compared with single photovoltaic or photothermal systems, the system's instantaneous comprehensive energy efficiency ratio is significantly improved, and the energy output of the solar energy collection device per unit area is significantly increased, greatly enhancing the comprehensive utilization efficiency of solar energy resources.

[0028] 2. Continuous active heating, completely solving the problem of heating interruptions at night and during inclement weather.

[0029] The system innovatively integrates two modes: "solar thermal circulation heating" and "resistance heating (heating wire)," and is equipped with a battery energy storage unit. The intelligent control unit automatically switches the operating mode based on the antifreeze temperature and battery charge: during the day or when there is sufficient sunlight, it uses efficient solar thermal circulation heating; at night, on cloudy days, or when there is insufficient sunlight, it automatically switches to resistance heating mode, powered by the battery for continuous heating. This mechanism achieves continuous day and night heating, complementing sunny and cloudy conditions, and completely overcomes the shortcomings of traditional solar thermal systems that fail in the absence of sunlight and pure photovoltaic systems that rely on large-capacity energy storage, significantly improving the reliability of frost damage prevention for railway subgrades in cold regions.

[0030] 3. Achieve energy autonomy and off-grid operation, reducing dependence on external sources.

[0031] The system uses photovoltaic power generation to meet all the electricity needs of its circulating pumps, heating wires, control modules, etc., and is equipped with batteries for energy transfer in time and space, forming a completely self-sufficient off-grid heating system. It does not require connection to an external power grid, making it particularly suitable for remote, cold-region railway lines without grid coverage, significantly reducing the construction cost and maintenance difficulty of external power supply facilities.

[0032] 4. High degree of intelligence, enabling unattended adaptive control.

[0033] The remote intelligent control module incorporates dual-mode switching logic based on temperature thresholds, dynamic hysteresis algorithms, fault self-diagnosis, energy efficiency optimization, and thermal balance linkage strategies. The system can sense its own energy status and roadbed heat demand in real time, automatically select the optimal operating mode, and adaptively adjust control parameters (such as preheating and protection power thresholds) based on weather forecast data. Furthermore, the system supports 4G / 5G or satellite remote communication, enabling it to report status to the monitoring center and receive parameter updates, achieving intelligent management with minimal or even no human intervention.

[0034] 5. Provide precise, on-demand heating to avoid energy waste.

[0035] Based on the heat balance equation, a roadbed heating demand model is established. The system can calculate the required heating in real time according to parameters such as the actual roadbed temperature, frost damage prevention volume, and soil thermal properties. It also precisely controls the duty cycle of the heating wire or the speed of the circulating pump through pulse width modulation to match the actual heating with the demand. A multi-objective optimization function comprehensively considers energy efficiency ratio, heating matching degree, and battery discharge rate, further improving the economy and effectiveness of energy utilization.

[0036] 6. The system is economical and widely applicable, with low total life-cycle costs.

[0037] The system is independent of the power grid, requires no large-capacity batteries (energy storage needs are reduced through photothermal-resistance synergy), and reduces external energy procurement costs. Simultaneously, its integrated design (photovoltaic-thermal integration) saves installation space and equipment costs. Intelligent control strategies extend battery life (health management, temperature protection), reducing maintenance frequency and replacement costs. Compared to complex systems such as ground source heat pumps, this invention has a relatively simple structure, lower initial installation costs, and easier maintenance, making it promising for both new railway construction and existing line renovation in cold regions.

[0038] 7. Strong fault diagnosis and fault tolerance capabilities, and high operational reliability.

[0039] The system incorporates fault detection logic for multiple components, including temperature sensors, circulating pumps, and heating wires. When a sensor fault is detected, it automatically switches to a virtual sensor model to estimate the temperature; when a circulating pump fault is detected, it automatically switches to pure resistance heating mode, ensuring emergency heating capacity under critical fault conditions. Simultaneously, a machine learning-based fault prediction model can provide early warnings of equipment aging or performance degradation, guiding preventative maintenance and further enhancing the system's long-term reliability.

[0040] 8. Remote adaptive calibration and continuous optimization capabilities

[0041] The cloud-based monitoring platform utilizes Bayesian optimization algorithms to automatically search for optimal control parameters (temperature threshold, hysteresis, protection power, etc.) based on historical operational data from each roadbed section, and remotely sends updates to the on-site intelligent control modules. This enables the system to adapt to different climatic conditions, geological features, and operational requirements, achieving continuous self-optimization and always maintaining optimal operating status. Attached Figure Description

[0042] Figure 1 System overall layout diagram;

[0043] Figure 2 Detailed diagram of the internal workings and thermal circulation of the control system cabinet;

[0044] Figure 3 Schematic diagram of the side structure of the photovoltaic-photothermal composite array. Detailed Implementation

[0045] Example 1

[0046] This embodiment provides a collaborative power supply system for preventing frost damage to railway subgrade in cold regions. The system is applied to a test section of a railway in a cold region, where the lowest winter temperature can reach -40℃, the frost heave depth of the subgrade is approximately 2.5m, and the volume of frost damage prevention is approximately 120m³. Figure 1 As shown (see attached diagram in the manual), the system is arranged along the railway line on both sides of the roadbed slope and shoulder open space. The diagram shows the overall layout and relative positions of the rails, ballast, roadbed, photovoltaic support, photovoltaic-thermal composite array, photovoltaic power lines, and heating rods buried in the roadbed.

[0047] The photovoltaic-thermal composite array unit consists of multiple photovoltaic panels and a photothermal liquid storage unit integrated beneath each photovoltaic panel. The photovoltaic panels use monocrystalline silicon photovoltaic modules with a photoelectric conversion efficiency of 21%. The photothermal liquid storage unit is a sealed aluminum alloy cavity mounted flush against the back of the photovoltaic panel. The cavity is filled with antifreeze (a 50% ethylene glycol aqueous solution with a freezing point as low as -35°C). Thermally conductive silicone grease is coated between the upper surface of this cavity and the back of the photovoltaic panel to reduce contact thermal resistance. Its working principle is as follows: When sunlight shines on the front of the photovoltaic panel, some photons are absorbed by the photovoltaic cells and converted into direct current (DC) energy. The remaining photons, including unused solar radiation (mainly in the near-infrared band) and waste heat generated by Joule heating during photovoltaic operation, are conducted to the photothermal liquid storage unit through the back of the photovoltaic panel. The antifreeze absorbs this heat, increasing its temperature and achieving photothermal conversion. At the same time, the flow of antifreeze carries away the heat from the back of the photovoltaic panel, reducing the operating temperature of the photovoltaic panel by about 15-20°C, thereby increasing its power generation efficiency by about 8%-12% compared to the uncooled state. This "dual-purpose" structural design outputs both electrical and thermal energy within the same light-receiving area, increasing the comprehensive utilization rate of solar energy per unit area from about 20% for photovoltaic alone or about 50% for solar thermal alone to over 70%.

[0048] The heat transfer and circulation unit includes heating rods, a circulation pump, an upward duct, and a downward duct, all embedded in the roadbed. The heating rods are seamless 316L stainless steel pipes with an inner diameter of 40mm and a wall thickness of 3mm. One rod is placed every 2m along the longitudinal direction of the roadbed, each 3m long, and horizontally buried 1.2m below the surface of the roadbed, within the seasonal freeze-thaw activity layer. The heating rods have hollow interiors, with both ends welded to the upward and downward ducts, respectively. The circulation pump is a low-temperature resistant shielded centrifugal pump with a rated head of 8m and a rated flow rate of 2.5m³ / h, installed in the control system cabinet. Both the upward and downward ducts are polyethylene composite pipes with an external insulation layer of 30mm thickness and a thermal conductivity not exceeding 0.03W / (m·K). The photothermal fluid storage unit, the upward duct, the heating rods, and the downward duct are connected by pipelines to form a closed circulation loop filled with antifreeze. When the circulation pump starts, the antifreeze is pumped from the upward duct into the heating rod. As it flows through the heating rod, it transfers heat to the surrounding subgrade soil through the pipe wall. The cooled antifreeze returns to the photothermal liquid storage unit through the downward duct and is heated again, thus repeating the cycle.

[0049] The energy storage and power supply unit includes a photovoltaic inverter / controller unit, a battery bank, and cables laid to the heating rods. The photovoltaic inverter / controller unit integrates an MPPT solar controller and a pure sine wave inverter with a rated power of 3kW. It can perform maximum power point tracking, voltage regulation, and rectification of the DC power (nominal voltage 48V) output from the photovoltaic panels, and manage the charging and discharging of the battery bank. The battery bank consists of four 12V / 200Ah lithium iron phosphate batteries connected in series, with a total nominal voltage of 48V, a rated capacity of 9.6kWh, and an operating temperature range of -20℃ to 60℃. It is also equipped with a battery heating pad. The cables are cold-resistant RVVP shielded cables, which are led out from the battery bank and laid along the roadside slope, ultimately connecting to the heating wires integrated within the heating rods. The heating wire is a low-power, high-heat-dissipation nano-heating core, specifically a carbon nanotube thin film heating element with a thickness of only 0.5mm, an operating voltage of 48V, a rated power of 600W, an electrothermal conversion efficiency of over 99%, and a positive temperature coefficient characteristic, which can automatically limit the temperature and avoid overheating.

[0050] The core of the intelligent control unit consists of a PT100 platinum resistance temperature probe (accuracy ±0.2℃) housed within the photothermal fluid storage unit and a remote intelligent control module installed in the control system cabinet. This control module is based on an embedded microcontroller with an ARM Cortex-M7 core, incorporates a real-time operating system, and integrates data acquisition, logic operations, control output, and 4G / 5G remote communication functions. Figure 2As shown (please refer to the attached diagram in the instruction manual), the control system cabinet also integrates a photovoltaic inverter control unit, a circulating pump, and a remote intelligent control module. It also shows the circulation loop formed by the photovoltaic thermal liquid through the upward and downward conduits, the heating rod, and the photovoltaic thermal liquid storage unit, as well as the logical relationship of the intelligent control module's control over the circulating pump and the heating wire. Figure 3 The cross-section shows the photovoltaic panel, the underlying photothermal liquid storage unit, and its internal heat medium, intuitively demonstrating the integrated structure of photovoltaic power generation and photothermal collection.

[0051] The core control strategy of the control module is a dual-mode switching based on temperature threshold and hysteresis logic, combined with battery power management. Specifically, the temperature probe collects the antifreeze temperature in real time. The control module reads the temperature value every 30 seconds. When sufficient sunlight is detected, the system activates the circulation pump, shuts off the heating element, and enters the solar thermal circulation heating mode. At this time, the antifreeze, driven by the circulation pump, flows through the solar thermal liquid storage unit to absorb heat, and then enters the heating rods in the roadbed through the upward duct, transferring heat to the roadbed soil. The cooled antifreeze returns through the downward duct, forming a closed thermal cycle. Because solar thermal conversion directly converts solar radiation energy into heat energy, and the circulation pump consumes only about 50W, while the solar thermal output power can reach over 1.5kW, the energy efficiency ratio (output heat / input electrical energy) of this mode is as high as 30 or more, making it extremely economical. hour( (This is the temperature hysteresis, defaulted to 2℃), and the battery pack capacity. If the battery's energy is insufficient, the control module shuts down the circulating pump and starts the heating element, putting the system into resistance heating mode. In resistance heating mode, the electrical energy stored in the battery is directly converted into Joule heat through a nano-heating core, and the surface temperature of the heating rod can reach 50℃~70℃, continuously supplying heat to the roadbed. Although this mode has an energy efficiency ratio of 1 (i.e., 1 unit of electricity produces 1 unit of heat), its advantage lies in being completely unaffected by sunlight conditions, allowing it to operate continuously at night or on cloudy days. If the battery pack's charge... If the system enters protection mode, it will shut down all heating equipment and issue a low battery alarm, waiting for photovoltaic charging to resume the next day. A hysteresis of 2℃ is set (i.e.,...). When the temperature drops from above 30°C, it needs to drop to 28°C before switching to electric heating; when the temperature rises from below 28°C, it needs to rise to 30°C before switching back to photothermal cycle. This effectively avoids frequent mode switching caused by small temperature fluctuations and prevents premature aging of the circulation pump and heating wire contact relay.

[0052] In the above control logic, the 30℃ threshold is not an arbitrary value, but a comprehensive optimization result based on the thermal requirements of railway subgrade in cold regions and the safety margin of antifreeze freezing point. Experimental results show that when the antifreeze temperature is below 25℃, the heat flux density transferred to the subgrade through the heating rods decreases significantly (due to the reduced temperature difference), making it difficult to effectively suppress frost heave under extreme low-temperature conditions. Conversely, setting the threshold too high (e.g., 40℃) would force the system to use electric heating most of the time, increasing battery wear. The 30℃ threshold, verified through simulation and field tests, can fully utilize solar heat in most sunny weather conditions while ensuring sufficient battery capacity when switching to electric heating at night. Similarly, the minimum charge protection threshold of 20% is determined based on the relationship curve between the depth of discharge and cycle life of lithium iron phosphate batteries: a discharge depth exceeding 80% (i.e., SOC below 20%) accelerates battery capacity decay. Setting the protection point at 20% allows the battery cycle life to reach over 3000 cycles, meeting the design service life of over 15 years for railways in cold regions.

[0053] To ensure the system's reliability and intelligence during long-term operation, the control module incorporates system initialization and self-test logic. Upon system power-on or reset, the control module sequentially executes the following initialization steps: First, it reads the initial value from the temperature probe to determine if it is within the physically valid range. If the system exceeds this range, it locks the system and sends a sensor fault alarm to prevent erroneous control due to sensor failure. Then, it reads the initial battery voltage and estimates the remaining charge. If the charge is below the minimum protection threshold of 20%, the system enters standby charging mode, not starting any heating equipment. Heating can only resume when the photovoltaic power is charged to 25% or higher. Next, it checks the offline status registers of the circulating pump and heating wire. If there are unresolved faults from the previous operation (such as pump jamming or heating wire overcurrent protection), the fault lockout state is maintained until manual reset. Afterward, the communication module is initialized, sending a system startup success message to the remote monitoring center, including the device ID, initial status parameters, and timestamp. After completing the above self-checks, the system enters the main control loop.

[0054] The main control loop executes a series of tasks periodically at fixed 30-second intervals. This cycle is chosen because the subgrade soil has a relatively large thermal inertia (temperature change time constant is approximately 5–10 minutes), and a 30-second sampling and control cycle is sufficient to capture temperature change trends without incurring unnecessary computational and communication overhead. Within each cycle, the control module first reads the current antifreeze temperature. The system determines the battery's state of charge (SOC) and then executes the aforementioned dual-mode switching control. Simultaneously, it performs fault self-diagnosis and alarm functions. More importantly, the control module calculates the system's instantaneous comprehensive energy efficiency ratio every 10 minutes (i.e., every 20 control cycles). The calculation formula is:

[0055] in, The heat absorbed by the photothermal fluid storage unit per unit time (unit: kW) is measured by measuring the temperature difference between the antifreeze entering and exiting the photothermal fluid storage unit. and traffic Calculated, i.e. , The specific heat capacity of the antifreeze is approximately 3.3 kJ / (kg·K). Density (approximately 1050 kg / m³); G is the electrical power output of the photovoltaic panel (unit: kW), directly measured by the photovoltaic inverter control unit; G is the solar irradiance (unit: kW / m²), measured by the total radiation meter installed next to the photovoltaic array; A is the effective light-receiving area of ​​the photovoltaic-photothermal composite array (unit: m²). This energy efficiency ratio... This reflects the system's overall utilization efficiency of solar irradiance energy, and its value is typically between 0.5 and 0.8. With the preset optimal energy efficiency range In comparison, in this embodiment .like This indicates that the current system operating efficiency is low, possibly due to excessively high circulating pump speed leading to excessive pump power consumption (in solar thermal mode) or mismatch in the duty cycle of the heating wire (in resistance heating mode). In this case, an adjustment operation is triggered: in solar thermal circulation mode, the control module reduces the circulating pump speed from the rated speed to 60% of the rated speed via the frequency converter, and the pump power decreases from 50W to approximately 18W, while the solar thermal absorption... The decrease in heat transfer coefficient due to reduced flow velocity is less than 10%, therefore This improves efficiency; in resistance heating mode, pulse width modulation (PWM) is used to control the equivalent heating power of the heating wire, adjusting the duty cycle from 100% to 60%–80%, keeping the battery discharge current within the optimal economic range (0.2C–0.5C) and reducing battery internal resistance loss. If the current operating parameters remain unchanged, the introduction of this energy efficiency optimization algorithm improves the system's daily average energy efficiency ratio by approximately 12% to 15% under different irradiation conditions.

[0056] To more accurately match the actual thermal demand of the roadbed, the control module also incorporates a roadbed heat supply demand model based on thermal balance. The physical basis of this model is that the heat required for a change in the temperature of the roadbed soil is equal to the increase in its internal energy. The roadbed thermal balance equation is established as follows:

[0057]

[0058] In the formula, The specific heat capacity of the subgrade soil is taken as 1.2 kJ / (kg·K) for silty clay, which is common in cold regions. The density of the soil is taken as 1.9 × 10³ kg / m³; The volume for frost damage prevention is determined based on the roadbed cross-sectional dimensions and frost depth range; in this embodiment, it is 120 m³. The target control temperature was set at -1℃ (studies have shown that frost heave is significantly aggravated when the temperature of the subgrade soil is below -1.5℃, and controlling it above -1℃ can effectively suppress frost heave). The real-time roadbed temperature is obtained by averaging the values ​​from auxiliary temperature sensors (one every 5m) buried around the heating rods, in °C. The heating time is set at 3600 seconds (1 hour) when calculating instantaneous heating demand. The control module calculates the required heating demand every 30 minutes. The unit is kW. In resistance heating mode, the control module... The formula for dynamically adjusting the PWM duty cycle D of the heating wire is:

[0059] In the formula, This refers to the real-time voltage of the battery (which varies with discharge, ranging from 40V to 54V). This refers to the rated current of the heating wire (12.5A, corresponding to a rated power of 600W / 48V). For example, when calculated... , ,but This means that the heating wire intermittently heats at a 50% duty cycle. This on-demand heating method avoids energy waste and roadbed overheating caused by prolonged full-power heating of the heating wire (excessive melting may cause frost heave and mudslides). Through comparative tests, after adjusting the duty cycle using a thermal balance model, the average nighttime battery power consumption decreased by 28%, while the roadbed temperature remained stable within the target range of -1℃ to 0℃.

[0060] It is worth emphasizing that in the above heat balance equation Take 3600 seconds. The use of parameters such as -1℃ is not common knowledge, but rather an optimized value obtained by the applicant through extensive numerical simulations and field tests of railway subgrade temperature fields in cold regions. Using other common simplified models, such as setting a fixed duty cycle based solely on ambient temperature (e.g., heating the heating wire at full power when the ambient temperature is below -20℃), would lead to two adverse consequences: first, overheating and energy waste when the initial subgrade temperature is high; and second, insufficient heating when the subgrade volume is large or the soil heat capacity is high. The heat balance model used in this embodiment, based on real-time subgrade temperature feedback and soil thermal properties, can adapt to different road sections, seasons, and initial temperature conditions, significantly improving the accuracy and energy efficiency of heating.

[0061] In addition to real-time control, the system also features adaptive predictive control, enabling heating behavior to "anticipate" future weather changes. The control module collects and stores system operation data for at least the past 7 days, including daily cumulative solar irradiance, minimum nighttime ambient temperature, daily antifreeze temperature variation curve, and daily battery charge. Exponential smoothing (smoothing coefficient) is employed. Predict the solar irradiance and ambient temperature for each hour of the next 24 hours. The recursive formula for exponential smoothing is: ,in The actual observed value at the current moment. This is a predicted value. The advantages of this method are low computational cost, suitability for embedded real-time systems, and better capture of recent trends compared to a simple moving average. Control parameters are dynamically adjusted based on the prediction results: if the predicted next day is sunny (peak solar irradiance...). If the nighttime discharge is allowed to reach 15% of the rated capacity (i.e., the SOC lower limit is lowered from 20% to 15%), it can be quickly recharged and restored during the day the following day; if the next day is predicted to be cloudy or snowy (peak day), then the discharge is allowed to reach 15% of the rated capacity (i.e., the SOC lower limit is lowered from 20% to 15%). If this is the case, then the protection power threshold will be raised to 30% of the rated capacity in advance, and the temperature hysteresis range of the photothermal cycling mode will be shortened (to...). The temperature was reduced from 2°C to 1°C to prioritize solar thermal energy use and reduce reliance on electric heating wires when sunlight is limited. This predictive control strategy reduced the probability of battery depletion by more than 40% during consecutive cloudy days, effectively ensuring heating continuity under extreme weather conditions.

[0062] Furthermore, the control module also implements fault self-diagnosis and alarm functions, including the detection of three typical faults. First, temperature sensor fault detection: when the read temperature value exceeds the physically possible range... If the sensor malfunctions, the system automatically switches to a virtual sensor model based on historical temperature and irradiance data to estimate the antifreeze temperature. The virtual sensor model uses multiple linear regression. Where G is the real-time solar irradiance, The ambient temperature is used, and coefficients a, b, and c are obtained through offline fitting using normal data from the past 7 days. Tests show that within 48 hours of sensor failure, the virtual sensor estimation error does not exceed 2°C, sufficient to maintain basic system operation until repair. Second, circulating pump failure detection: When the circulating pump is running but the antifreeze temperature remains above 30°C for more than 5°C (i.e.,... If the temperature shows no significant downward trend (temperature drop < 1℃ within 10 minutes), it is determined that the circulating pump may be faulty (e.g., impeller jamming, pipeline blockage). In this case, the system first attempts a backflush operation (short-term reverse pump rotation) 3 times. If this is ineffective, the photothermal circulation is stopped and an alarm is sent. Third, abnormal heating wire efficiency detection: When the heating wire is in the start-up state but the battery pack charge drops by more than 15% of the rated capacity within 10 minutes, it is determined that the heating wire efficiency is abnormal (possibly due to poor contact leading to increased resistance, and the actual heating power is lower than the set value). When any of the above faults occur, the control module sends an alarm message containing the fault type, timestamp, and suggested handling measures to the remote monitoring center via 4G / 5G or satellite communication.

[0063] Furthermore, to further improve energy efficiency, the control module employs a more refined energy efficiency optimization algorithm. Its core principle is to adjust actuator parameters under different operating modes to achieve optimal performance. In the solar thermal cycle mode, the circulating pump does not always operate at its rated speed, but rather adjusts its speed based on the real-time energy efficiency ratio. Adjustments are made. This embodiment establishes the calculation basis for the optimal speed of the circulating pump through theoretical derivation: under the photothermal cycle mode, the overall system efficiency is... ,in This refers to the electrical power of the circulating pump. It increases with increasing flow rate, but the rate of increase gradually slows down (because heat exchange tends to saturate). It is proportional to the cube of the rotational speed. Therefore, there exists an optimal rotational speed. Make Maximum. Experimentally calibrated, this system... That is, 60% of the rated speed. When Below At that time, the control module directly sets the speed to Rather than simple linear adjustment, it achieves optimal performance in one step. In resistance heating mode, energy efficiency optimization is mainly reflected in the dynamic adjustment of the PWM duty cycle, while also considering the discharge efficiency characteristics of the battery: lithium iron phosphate batteries have the highest energy efficiency (approximately 95%) at discharge rates of 0.2C to 0.5C; excessively high or low rates will lead to increased internal resistance losses. Therefore, the control module must meet the following requirements... Under the premise of ensuring that the duty cycle of the heating wire is adjusted to make the discharge current around 0.3C, if If the demand is small, intermittent full-power heating (e.g., heating for 5 minutes and stopping for 10 minutes) is used to avoid inefficient low-current discharge.

[0064] The working process of the system described in this embodiment can be summarized as follows: During the day when there is sunshine, the photovoltaic-solar thermal composite array generates electricity and heat simultaneously. The photovoltaic power prioritizes charging the battery, while the solar thermal power drives the antifreeze circulation to heat the roadbed via a circulating pump. When the antifreeze temperature is below 28°C and the battery has sufficient charge, it automatically switches to heating wire heating. When the battery charge is below 20%, it enters protection mode. Throughout the process, the control module collects data every 30 seconds, performs energy efficiency optimization calculations every 10 minutes, adjusts the heating demand based on the actual roadbed temperature every 30 minutes, and uses weather forecast information to adjust the control strategy in advance. The system uploads status data to a cloud monitoring platform via a 4G / 5G network, and the platform can remotely issue parameter updates. After three consecutive months of field operation testing during winter, the temperature at a depth of 1.5m below the top surface of the roadbed in the test section using this embodiment of the system remained consistently between -1.2℃ and 0.5℃, while the temperature at the same depth in the control section without the prevention and control system was as low as -5.8℃. The frost heave was reduced from 12mm in the control section to less than 2mm, fully meeting the requirements for railway line smoothness. The system provides an average daily heat supply of 8.2kWh, of which solar thermal energy contributes approximately 6.1kWh (74%), electric heating wire contributes 2.1kWh (26%), and photovoltaic power generation totals 5.8kWh. After deducting the power consumption of the circulating pump, the average daily net discharge of the battery is only 0.5kWh, achieving a high degree of energy self-sufficiency.

[0065] Example 2

[0066] This embodiment provides an operational method for preventing frost damage to railway subgrade in cold regions based on the system described in Embodiment 1. This method is automatically executed by an embedded program within a remote intelligent control module, aiming to achieve adaptive and intelligent operation of the system under all-weather, multi-season, and different meteorological conditions, maximizing energy utilization efficiency and heating reliability. The following detailed description of the method, along with specific operational steps, further illustrates this method.

[0067] After the system is powered on or reset, it first performs initialization and self-test steps. The control module reads the initial value from the temperature probe. Determine whether it is within the physical effective range. Within this range. This range is selected based on the linear measurement range of the platinum resistance temperature sensor and the extreme temperatures that may occur in cold environments. If the temperature exceeds this range, it indicates that the sensor is open-circuited, short-circuited, or damaged. The control module will immediately lock the system and send a sensor fault alarm to avoid misjudgments due to incorrect temperature values ​​(e.g., misjudging an extremely low temperature as a high temperature and stopping heating). Next, the initial voltage of the battery pack is read. The remaining power is estimated by looking up a table. The open-circuit voltage of a lithium iron phosphate battery has a monotonic relationship with its state of charge (SOC), and its accuracy can reach within 5% under static conditions. If... The system enters standby charging mode, without activating any heating equipment, and illuminates the charging indicator light. This is because activating the heating element when the battery level is too low can cause over-discharge of the battery, permanently damaging it. Once the photovoltaic charging raises the SOC to above 25%, the system automatically exits standby mode. Then, it checks the offline status registers of the circulation pump and heating element. If any unrecovered faults exist from the previous operation (such as circulation pump overcurrent protection lockout, heating element short-circuit flag set to 1), the fault lockout status is maintained, and a maintenance request is reported until manually reset. Finally, the 4G / 5G communication module is initialized, sending a system startup success message to the remote monitoring center, including the device ID, initial temperature, initial battery level, and current timestamp. After completing the above self-checks, the system enters the main control loop.

[0068] The main control loop executes periodically at fixed time intervals of 30 seconds. This cycle is based on the thermal diffusivity of the subgrade soil. Certainly, among them The thermal conductivity of the soil is approximately 1.5 W / (m·K). The volumetric heat capacity is approximately 2.28 × 10⁻⁶. 6 J / (m³·K)), calculated to The characteristic time for a temperature disturbance to propagate 1 meter is approximately... The estimated time is approximately 17 days, but in reality, the temperature response around the heating element is much faster (due to direct contact with the heat source), with a time constant of about 5-10 minutes. A 30-second sampling period is much shorter than this time constant, satisfying Shannon's sampling theorem while avoiding excessively frequent control switching. Within each cycle, the control module reads the current antifreeze temperature. Check the battery's state of charge (SOC) and then perform the mode switching procedure.

[0069] The mode switching step employs a dynamic hysteresis algorithm to overcome the insufficient adaptability of a fixed hysteresis value under different temperature change rates. Dynamic hysteresis value Based on the rate of temperature change Adjust in real time. The calculation uses the backward difference method: ,in .when When the temperature changes by more than 0.6°C per minute, it indicates drastic weather changes (such as rapid cloud movement or sunrise / sunset). The temperature was reduced from the default 2°C to 1°C to speed up mode switching response and prevent excessively long heating interruptions; when At that time, the temperature changes gradually, The hysteresis is increased to 3°C to prevent frequent switching caused by measurement noise or minor fluctuations. The theoretical basis of this dynamic hysteresis algorithm is that during rapid changes, the system state quickly crosses the threshold range; reducing the hysteresis will not significantly increase the number of switching operations, but it can reduce response delay. In a steady state, temperature fluctuations are mainly caused by sensor noise and control errors; increasing the hysteresis can effectively filter out these interferences. Simulation comparisons show that using the dynamic hysteresis algorithm, compared to a fixed 2°C hysteresis, reduces the number of invalid switching operations by approximately 35% in alternating sunny and cloudy weather, while shortening the nighttime switching delay by approximately 40%. Then, mode switching is performed according to the following logic: If... Start the circulation pump, turn off the heating element, and enter the solar thermal circulation heating mode; if and Turn off the circulation pump, start the heating element, and enter resistance heating mode; if Turn off all heating equipment and enter protection mode.

[0070] Building upon mode switching, the system also features an adaptive prediction pre-switching step. This step utilizes the adaptive prediction control function (exponential smoothing method for predicting future weather) described in Example 1 to proactively intervene before mode switching occurs. Specifically, if the solar irradiance G is predicted to decrease to a certain level within the next hour... When the weather is near sunset or covered by heavy clouds, and the system is currently in solar thermal cycling mode, the control module will preheat the heating element to raise the antifreeze temperature. Maintain a temperature above 25°C. The reason for choosing 25°C is that at this temperature, the antifreeze has sufficient heat capacity to continue providing heat for a period after switching, relying on residual heat and reducing the direct operating time of the heating element. The preheating power is determined based on heat balance. ,in The specific heat capacity of antifreeze (unit: J / (kg⋅K)). This represents the total mass of antifreeze in the circuit (in kg). The remaining solar thermal time is estimated (in seconds, calculated based on the rate of decrease in solar altitude angle). If consecutive cloudy days are predicted to exceed 48 hours, the system will be forcibly switched to energy-saving mode: the heating wire activation threshold will be lowered from 30°C to 25°C, and the battery protection charge will be increased from 20% to 30%. The reason for lowering the activation threshold is that solar thermal contribution is minimal during consecutive cloudy days, and the antifreeze temperature is unlikely to reach 30°C. Maintaining the original threshold would result in the heating wires never activating, leaving the roadbed without heating. Increasing the protection charge is to reserve more energy to cope with longer periods of cloudy weather. This pre-switching strategy increases the system's heating durability under continuous severe weather conditions from 68% to 94%.

[0071] Fault self-diagnosis and fault tolerance procedures are executed within each control cycle. The control module sequentially detects three types of anomalies: temperature sensor malfunction, circulating pump malfunction, and heating wire efficiency malfunction. When a temperature sensor malfunction is detected (reading value exceeds...), the module will automatically resolve the issue. When this occurs, the system automatically switches to a virtual sensor model to estimate the antifreeze temperature. The virtual sensor model uses a multiple linear regression approach: Where G is solar irradiance, The ambient temperature is represented by coefficients a, b, and c, which were obtained offline using the least squares method based on measured data from the past 7 days of normal operation. The physical meaning of this model is that the antifreeze temperature is primarily determined by solar radiation heating and ambient cooling. When a sensor malfunctions, the control module calls this model every 30 seconds to estimate the temperature, replacing the actual sensor value for mode switching. Tests show that over 48 hours, the average estimation error is approximately 1.2℃, with a maximum error of 2.8℃, still ensuring basic system operation. When a circulation pump malfunction is detected (during operation...),... If the temperature drop is less than 1°C within 10 minutes, the system first performs three backflush operations: controlling the circulating pump to rotate in the reverse direction for 3 seconds, then in the forward direction for 5 seconds, repeating this cycle 3 times. Backflush removes impurities adhering to the impeller or breaks up air bubbles. If this is ineffective, it is considered an unrecoverable fault, automatically switching to pure resistance heating mode and increasing the duty cycle of the heating element to compensate for heat loss: the compensated duty cycle... This means that the heating power can be increased by up to 30%. When abnormal heating wire efficiency is detected (power drop of more than 15% of rated capacity within 10 minutes during startup), the control module issues an alarm and reduces the duty cycle to 50% to prevent further overcurrent damage.

[0072] The energy efficiency optimization and thermal balance linkage step is executed every 10 minutes. This step first calculates the system's instantaneous comprehensive energy efficiency ratio according to the formula in Example 1. and the heat required for the roadbed Then construct the multi-objective optimization function J:

[0073]

[0074] in, For the weighting coefficient, in this embodiment, we take... ; The theoretical maximum energy efficiency ratio is taken as 0.85 (the upper limit after considering optical and thermal losses). To provide actual heat supply, in the solar thermal mode, through Calculations were performed in resistance heating mode. calculate; The battery discharge rate is defined as... The unit is % / h. The physical meaning of this optimization function is to suppress excessively fast battery discharge while maximizing the energy efficiency ratio and minimizing heating deviation (because high current discharge reduces battery energy efficiency and shortens lifespan). The control module determines the optimal combination of circulating pump speed and heating wire duty cycle through online optimization. The optimization adopts a simplified version of the gradient descent method: in solar heating mode, the circulating pump speed is changed in 10% steps (from 50% to 100% of rated speed), the corresponding J value is calculated, and the speed that maximizes J is selected; in resistance heating mode, the duty cycle is changed in 5% steps (from 20% to 100%), and the optimal duty cycle is selected. Since the optimization is only performed once every 10 minutes, the computational overhead is acceptable. Compared with fixed parameter operation without optimization, this multi-objective optimization improves the daily average energy efficiency ratio by 18% and the battery discharge depth consistency by 25%.

[0075] The on-demand precision heating process continues in resistance heating mode. The control module outputs the optimal duty cycle based on a multi-objective optimization function. The on / off time ratio of the heating wire is controlled by a pulse width modulation (PWM) signal. The PWM frequency is set to 1Hz (1-second period) to avoid temperature fluctuations caused by insufficient thermal inertia of the heating wire. Actual heating power... To ensure accurate matching of actual heat supply The control module incorporates proportional feedback correction: it monitors the feedback from the internal temperature sensor of the roadbed every 5 minutes to calculate the measured temperature rise rate. Compared with the theoretically expected rate of temperature rise The deviation between them. If the measured rate is lower than expected, increase the correction amount for the duty cycle:

[0076] in This is the proportional gain. The feedback controller is designed based on a first-order thermal system model, which can eliminate steady-state errors caused by uncertainties in soil parameters or changes in ambient temperature. For example, when... Actual measurement When the deviation is 1.5℃, then This means the duty cycle increases by 7.5%. After 3 to 4 feedback cycles (approximately 15 to 20 minutes), the roadbed temperature can converge to near the target value.

[0077] The battery health management process runs continuously to extend the battery pack's lifespan. The control module records the initial State of Charge (SOC), final SOC, and discharge duration for each discharge, and calculates the Depth of Discharge (DOD). The Depth of Discharge is defined as... When the cumulative depth of discharge exceeds a preset threshold When the battery is at its equivalent value of 100 cycles of 100% deep discharge, the control module assesses the battery health status (SOH). The SOH estimate is based on a discharge capacity test: with the system in standby mode and the battery fully charged, it is discharged at a constant current of 0.2C to the cutoff voltage, and the actual capacity discharged is recorded. With nominal capacity The ratio is SOH. When At this time, the control module dynamically adjusts the minimum protection charge: gradually increasing it from the original 20% to 30% (increasing the protection charge by 2% for every 5% decrease in SOH). This is because the internal resistance of aging batteries increases, making them more susceptible to voltage drops and accelerated capacity decay due to over-discharge. Simultaneously, the photothermal cycling mode is prioritized, reducing the frequency of heating wire usage to decrease battery charge-discharge cycles. Furthermore, when the temperature probe detects that the battery compartment temperature is below 0°C, the usable capacity of the lithium iron phosphate battery significantly decreases (capacity at -10°C is only 70% of that at 25°C), and its charging acceptance deteriorates. Therefore, the control module activates the heating wire to intermittently heat the battery compartment: when the battery temperature... Furthermore, when the system is in resistance heating mode, 50W is allocated from the total power of the heating wire to heat the battery compartment, continuously heating until... Then stop. This strategy increases the battery's discharge efficiency at low temperatures from approximately 65% ​​to over 90%.

[0078] The status reporting and remote optimization steps are executed every 30 seconds. The control module generates a status record, including a timestamp, The system records data including SOC, operating mode, circulating pump status, heating wire status, cumulative heat output (kWh), cumulative power generation (kWh), battery health status (SOH), and fault codes. These records are uploaded to a cloud monitoring platform via 4G / 5G or satellite communication. After receiving data from multiple roadbed sections, the cloud platform uses a time-series prediction model (such as the Seasonal Differential Autoregressive Moving Average (SARIMA) model) to continuously optimize the control parameters for the next day. The SARIMA model has an order of (p,d,q)×(P,D,Q)s, where p=2, d=1, q=2, the seasonal period s=24 (hours), and seasonal parameters P=1, D=1, Q=1. The model input consists of the hourly average solar irradiance and ambient temperature over the past 7 days, and the output is the predicted value for the next 24 hours. Based on the prediction results, the platform uses a Bayesian optimization algorithm to search for the optimal temperature threshold. Hysteresis Protect power Parameter combination. The objective function of Bayesian optimization is: ,in For parameter vectors, This is the average energy efficiency ratio over the past 7 days. To protect the trigger rate of mode, The penalty coefficient is set to 0.5. The optimized parameter set is updated remotely to the intelligent control modules of each road segment to achieve continuous self-optimization.

[0079] To further enhance the system's predictive maintenance capabilities, this embodiment also includes a fault prediction step based on machine learning. The control module collects at least 30 days of historical operating data and extracts the following features: the standard deviation of the temperature change rate. Energy efficiency ratio fluctuation range First-order difference mean of the battery discharge curve These characteristics reflect the health status of the system. For example, early signs of circulation pump jamming include a significant increase in the standard deviation of the temperature change rate (due to sudden temperature rises and falls caused by circulation interruption); aging of the heating wire is characterized by increased fluctuations in the energy efficiency ratio (due to unstable actual power caused by resistance changes); and battery performance degradation is manifested by the first-order difference mean of the discharge curve deviating from the normal range (due to a sudden voltage drop caused by increased internal resistance). An isolated forest anomaly detection model is adopted, which constructs multiple isolated trees by randomly cutting the feature space, making it easier to isolate anomalies (shorter path length). During model training, normal data is marked as 0, and artificially injected simulated fault data is marked as 1. During online operation, real-time features are input into the model. If the model outputs anomaly scores exceeding a threshold (set to 0.65 in this embodiment), an early warning of possible fault types is issued, and preventive maintenance is recommended (such as cleaning the circulation pump filter, replacing the heating wire, and equalizing charging). Testing showed that the model's average early warning time for circulation pump jamming is 48 hours, for heating wire aging is 72 hours, and for battery performance degradation is as long as 120 hours.

[0080] During transitional seasons or periods of day-night alternation, antifreeze temperatures often fluctuate between 28°C and 30°C, causing the system to frequently switch between photothermal and resistive modes. This frequent switching not only increases relay wear but also causes pressure surges in the piping due to frequent start-stop cycles of the circulation pump. To address this issue, this embodiment introduces a multi-mode collaborative optimization step. When If this continues for more than 5 minutes, the control module determines that it has entered a transition zone. At this point, instead of a simple two-way switching, it simultaneously operates the circulating pump and heating element at low power, forming a hybrid solar-thermal and resistance heating mode. Specifically, the circulating pump operates at 30% of its rated speed (pump power approximately 5W), and the heating element pulses at a 10% duty cycle (equivalent power 60W). The principle of the hybrid mode is: when solar-thermal contribution is insufficient but some heat can still be recovered, low-power electric heating is used as a supplement to maintain uninterrupted roadbed heating while avoiding mode switching. The total heat output in the hybrid mode is... ,in This represents the photothermal absorption at low flow rates. Experiments show that the system energy efficiency ratio in the hybrid mode is approximately 8–12, which is lower than that in the pure photothermal mode (>30), but much higher than that in the pure resistive mode (=1), and completely eliminates switching oscillations. When Once the temperature stabilizes above 30°C or below 28°C, the system exits the hybrid mode and resumes the regular dual-mode switching.

[0081] Finally, to achieve optimal system parameter configuration and adapt to different geographical and climatic conditions, this embodiment also includes a remote parameter adaptive calibration step. The cloud monitoring platform periodically (e.g., weekly) compares the system energy efficiency and frost damage prevention effect under different roadbed sections and different control parameters. Specifically, the platform calibrates each roadbed section according to the control parameter vector. Group the data and calculate the average energy efficiency ratio for each group. and the success rate of frost damage prevention (Defined as the percentage of time the roadbed temperature is above -2℃). Then, a Bayesian optimization algorithm is used to automatically search for the optimal parameter combination. The probabilistic surrogate model for Bayesian optimization uses a Gaussian process with a Matérn 5 / 2 kernel. The optimization objective is to maximize... ,in The weighting coefficient is 0.3. After multiple iterations (usually 5-10 rounds), the algorithm converges to the optimal parameters. The platform remotely updates the calibrated parameters to the intelligent control modules of each roadbed section. This continuous optimization mechanism enables the system to adapt to the climatic characteristics of different regions (such as the strong radiation and low temperature environment of the Qinghai-Tibet Plateau and the weak radiation and extreme cold environment of Northeast China), and can maintain optimal operating status for a long time without human intervention.

[0082] The operating method provided in this embodiment achieves efficient, reliable, and intelligent operation of the railway subgrade frost damage prevention system in cold regions through a series of technical means, including dynamic hysteresis switching, adaptive predictive pre-switching, fault tolerance, multi-objective energy efficiency optimization, on-demand precise heating, battery health management, machine learning fault prediction, multi-mode collaboration, and remote parameter self-calibration. Compared with the simple timed or temperature-controlled strategies in existing technologies, the method in this embodiment can dynamically optimize control parameters based on real-time environment, equipment status, and historical data, significantly improving energy utilization efficiency and heating continuity while reducing operation and maintenance costs. Actual line testing showed that the test section using this embodiment's method, compared to the control section using traditional temperature control (fixed threshold 30℃, fixed hysteresis 2℃, no prediction), achieved a 22% increase in daily energy efficiency ratio, a 30% extension in battery cycle life, and a 65% reduction in heating interruption time due to faults, fully verifying the technical superiority and high level of innovation of this embodiment.

[0083] This invention achieves synergistic energy supply from solar photovoltaic and solar thermal energy by constructing a photovoltaic-solar thermal composite array. It simultaneously outputs electrical and thermal energy within the same light-receiving area, significantly improving the overall utilization efficiency of solar energy. Through a dual-mode (solar thermal cycle and resistance heating) intelligent switching strategy and battery energy storage, the system can continuously and stably provide active heating to the roadbed under various weather conditions, including daytime, nighttime, and cloudy / snowy weather, completely solving the problem of failure of traditional solar heating systems when there is no sunshine. The intelligent control unit incorporates functions such as dynamic hysteresis switching, energy efficiency optimization, thermal balance demand matching, adaptive predictive control, and fault self-diagnosis, achieving unattended intelligent operation. This invention does not rely on an external power grid and is particularly suitable for railways in remote, cold regions. It has outstanding advantages such as energy self-sufficiency, reliable heating, good economy, and simple maintenance, providing a highly efficient, all-weather, adaptive active heating solution for the prevention of frost damage to railway roadbeds in cold regions, and has broad engineering application prospects.

[0084] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A collaborative energy supply system for preventing frost damage to railway subgrade in cold regions, characterized in that, include: A photovoltaic-photothermal composite array unit is installed on the slope or open space along a railway line to simultaneously output DC power and heat energy. The unit includes a photovoltaic panel and a photothermal liquid storage unit integrated below the photovoltaic panel. The photothermal liquid storage unit is filled with antifreeze to absorb solar radiation energy not utilized by the photovoltaic panel and waste heat generated when the photovoltaic panel is working, thereby realizing photothermal conversion and cooling the photovoltaic panel. The heat transfer and circulation unit includes a heating rod, a circulation pump, and an upward and downward conduit connecting the photothermal liquid storage unit and the heating rod, all buried in the subgrade bed. The heating rod has a hollow pipe inside for the antifreeze to flow through, and its outer wall is thermally conductive. The circulation pump is used to drive the antifreeze to circulate in a closed loop consisting of the photothermal liquid storage unit, the upward conduit, the heating rod, and the downward conduit, so as to transfer heat to the subgrade soil. The energy storage and power supply unit includes a photovoltaic inverter and controller, a battery bank, and cables laid to the heating rods; the photovoltaic inverter and controller is used to stabilize and rectify the DC power generated by the photovoltaic panels and manage the charging and discharging of the battery bank; the battery bank is used to store excess energy generated by photovoltaic power generation. The intelligent control unit includes a temperature probe installed in the photothermal liquid storage unit and a remote intelligent control module; the remote intelligent control module intelligently controls the start and stop of the circulation pump and the heating wire according to the temperature probe signal and the preset program, so as to realize the automatic switching of the working mode. The heating rod has an integrated heating wire inside or outside. The heating wire is electrically connected to the power supply unit via a cable and is used to convert electrical energy into heat energy. The heating wire is a nano-heating core with low power consumption and high heat dissipation.

2. The system according to claim 1, characterized in that, The remote intelligent control module is configured to execute a dual-mode switching control strategy based on temperature threshold and hysteresis logic: When the temperature probe detects the antifreeze temperature When the control circulation pump starts and the heating wire is turned off, the system enters the solar thermal circulation heating mode; When the antifreeze temperature At that time, if the battery pack power Then, the circulating pump is shut down, the heating element is started, and the system enters resistance heating mode; among which... This refers to the temperature hysteresis. If the battery pack capacity If this occurs, the system enters protection mode, shuts down all heating equipment, and issues a low battery alarm.

3. The system according to claim 1, characterized in that, The remote intelligent control module has built-in system initialization and self-test logic; when the system is powered on or reset, the control module executes the following initialization steps in sequence: Read the initial value of the temperature probe and determine if it is within the physical effective range. If the range is exceeded, the system will be locked and a sensor fault alarm will be sent. The system reads the initial voltage of the battery pack and estimates the remaining power. If the power is below the minimum protection threshold of 20%, the system enters standby charging mode and does not start any heating equipment. Check the offline status registers of the circulating pump and heating element. If there was an unrecovered fault during the last operation, keep the fault locked until manual reset. Initialize the communication module and send a system startup success message to the remote monitoring center, which includes the device ID, initial status parameters and timestamp; After completing the above self-check, the system enters the main control loop.

4. The system according to claim 1, characterized in that, The remote intelligent control module executes a main control loop, which periodically performs the following task sequence at fixed time intervals of 30 seconds: Read the current antifreeze temperature and battery SOC; Execute dual-mode switching control: If Then start the circulation pump and turn off the heating element; if Then shut down the circulation pump and start the heating element; if Then shut down all heating equipment and sound an alarm; Perform fault self-diagnosis and alarm functions; Perform system-wide energy efficiency optimization: calculate the system's instantaneous overall energy efficiency ratio every 10 minutes. The calculation formula is: ; in, The heat absorbed by the photothermal liquid storage unit per unit time, expressed in kW; G represents the electrical power output of the photovoltaic panel, in kW; G represents the solar irradiance, in kW / m²; A represents the effective light-collecting area of ​​the photovoltaic-photothermal composite array, in m². like Less than the lower limit of the preset optimal energy efficiency range In the photothermal cycle mode, the speed of the circulating pump is reduced to 60% of the rated speed, and in the resistance heating mode, pulse width modulation is used to control the equivalent heating power of the heating wire. Generate status records and report them to the monitoring center via remote communication; If an exception occurs during the cycle, the exception is captured and an emergency stop protection is executed, shutting down the circulation pump and heating wire, logging the error, and reporting it.

5. The system according to claim 1, characterized in that, The remote intelligent control module also performs fault self-diagnosis and alarm functions, including: Temperature sensor fault detection: When the read temperature value exceeds the physically possible range When this happens, it is determined that the sensor is malfunctioning; Circulation pump fault detection: When the circulation pump is running but the antifreeze temperature remains above [temperature value missing], Da If the temperature shows no obvious downward trend, the circulation pump is likely to be faulty. Detection of abnormal heating wire efficiency: When the heating wire is in the start-up state but the battery pack charge drops by more than 15% of the rated capacity within 10 minutes, it is determined that the heating wire efficiency is abnormal. When any of the above faults occurs, the control module sends an alarm message containing the fault type, timestamp, and suggested handling measures to the remote monitoring center via 4G / 5G or satellite communication.

6. The system according to claim 1, characterized in that, The remote intelligent control module also has a built-in system comprehensive energy efficiency optimization algorithm, which evaluates the system operating efficiency in real time and dynamically adjusts the control parameters based on the following formula: System instantaneous integrated energy efficiency ratio The calculation is as follows: ; The control module performs calculations every 10 minutes. and the preset optimal energy efficiency range Compare; in, The heat absorbed by the photothermal liquid storage unit per unit time, expressed in kW; G represents the electrical power output of the photovoltaic panel, in kW; G represents the solar irradiance, in kW / m²; A represents the effective light-collecting area of ​​the photovoltaic-photothermal composite array, in m². like If this is triggered, the following adjustment operations will be performed: in photothermal cycle mode, the speed of the circulation pump will be reduced to 60% of the rated speed; in resistance heating mode, pulse width modulation will be used to control the equivalent heating power of the heating wire. like If so, the current operating parameters will remain unchanged.

7. The system according to claim 1, characterized in that, The remote intelligent control module also has a built-in roadbed heating demand model based on thermal balance: Establish the roadbed heat balance equation: ; in, This refers to the specific heat capacity of the roadbed soil, expressed in J / (kg·K). This is the density of the soil, expressed in kg / m³. The volume for frost damage prevention is expressed in m³. The target temperature is specified in °C. This is the real-time roadbed temperature, in °C. Heating time is measured in seconds (s). The control module calculates the required heat supply every 30 minutes. ; In resistance heating mode, the control module according to The pulse width modulation duty cycle D of the heating wire is dynamically adjusted, and the calculation formula is as follows: ; in, This is the real-time voltage of the battery, in V. This is the rated current of the heating wire, expressed in amperes (A).

8. The system according to claim 1, characterized in that, The remote intelligent control module also has adaptive predictive control functionality: Collect and store system operation data for at least the past 7 days, including daily cumulative solar irradiance, minimum nighttime ambient temperature, daily antifreeze temperature variation curve, and battery end-of-day charge. Using the sliding window averaging or exponential smoothing method, predict the solar irradiance and ambient temperature for each hour in the next 24 hours; The control parameters are dynamically adjusted based on the forecast results: if the forecast is for a sunny day the next day, nighttime discharge is allowed up to 15% of the rated capacity; if the forecast is for a cloudy or snowy day the next day, the protection power threshold is raised to 30% of the rated capacity in advance, and the temperature hysteresis range of the solar thermal cycle mode is shortened.

9. A method for preventing and controlling frost damage to railway subgrade in cold regions based on the system described in any one of claims 1 to 8, characterized in that, Includes the following steps: System initialization and self-test steps: After the system is powered on, it performs temperature sensor validity detection, battery power detection, equipment fault status detection and communication initialization. After all of them pass, it enters the main control loop. Main control loop steps: Perform temperature reading, mode switching, fault detection, energy efficiency optimization and status reporting periodically at fixed time intervals of 30 seconds; Mode switching steps: Real-time monitoring of antifreeze temperature Based on the battery's state of charge (SOC), a dynamic hysteresis algorithm is used to calculate the switching threshold and the dynamic hysteresis amount. Based on the rate of temperature change Adjustment, unit ℃ / s: when hour, ;otherwise ;when When the time comes, start the circulation pump and turn off the heating element; when and When the time comes, turn off the circulation pump and start the heating element; when At that time, shut down all heating equipment; Adaptive prediction and pre-switching steps: If it is predicted that the solar irradiance will drop below 50 W / m² within the next hour and the current system is in photothermal cycle mode, the heating element will be activated in advance for preheating to maintain the antifreeze temperature above 25°C; if it is predicted that there will be more than 48 consecutive cloudy days in the future, the system will be forced to switch to energy-saving mode, lowering the heating element activation threshold to 25°C and increasing the battery protection capacity to 30%. Fault self-diagnosis and fault tolerance steps: Detect abnormal states of temperature sensor, circulation pump, and heating wire in each control cycle; when a temperature sensor fault is detected, automatically switch to a virtual sensor model based on historical temperature and irradiance to estimate antifreeze temperature; when a circulation pump fault is detected, automatically switch to pure resistance heating mode. Energy efficiency optimization and thermal balance linkage steps: Calculate the instantaneous comprehensive energy efficiency ratio of the system every 10 minutes. and the heat required for the roadbed Construct a multi-objective optimization function: ； in, These are the weighting coefficients. This is the theoretical maximum energy efficiency ratio. For actual heat supply, This refers to the battery discharge rate. The optimal combination of circulating pump speed and heating wire duty cycle is determined through online optimization. On-demand precise heating steps: In resistance heating mode, based on the optimal duty cycle The heating wire is controlled by a pulse width modulation signal; the actual heating power is: and make it match ; Battery health management steps: When the battery health level... When the battery temperature is below 0°C, the heating element will be activated to intermittently heat the battery compartment to above 5°C. Status reporting and remote optimization steps: A status record is generated every 30 seconds and uploaded to the cloud monitoring platform via wireless communication. The platform uses a time series prediction model to continuously optimize the control parameters for the next day based on the uploaded data and then distributes the results.

10. The method according to claim 9, characterized in that, It also includes the following steps: The fault prediction steps based on machine learning are as follows: Collect at least 30 days of historical operating data, extract features such as temperature change rate, energy efficiency ratio fluctuation, and battery discharge curve, and train an isolated forest anomaly detection model; when running online, input the real-time features into the model, and if the anomaly score exceeds the threshold, issue an early warning of possible circulation pump jamming, heating wire aging, or battery performance degradation. Multi-mode collaborative optimization steps: During the transitional season or the day-night cycle, when the antifreeze temperature fluctuates between 28°C and 30°C, the control module simultaneously operates the circulation pump and heating wire at low power. The circulation pump operates at 30% of its rated speed, and the heating wire pulses with a 10% duty cycle, forming a hybrid solar and resistance heating mode. Remote parameter adaptive calibration steps: The cloud monitoring platform regularly compares the system energy efficiency and frost damage prevention effect under different roadbed sections and different control parameters. It uses Bayesian optimization algorithm to automatically search for the optimal combination of temperature threshold, hysteresis and protection power parameters, and remotely updates the calibrated parameters to the intelligent control module of each roadbed section.

Images