Intelligent water replenishing control method and device for heating system

Abstract

The invention relates to the field of water replenishing control of heating systems, in particular to an intelligent water replenishing control method and device of a heating system. The method comprises the following steps: calculating a water supply and return temperature difference value and a standardized enthalpy difference sequence of a water supply pipeline and a water return pipeline of the heating system; collecting rated configuration information currently set by the heating system, and calculating a theoretical temperature difference reference value according to the rated configuration information; performing temperature difference deviation analysis on a theoretical temperature difference reference value according to the supply and return water temperature difference value and a standardized enthalpy difference sequence, and marking an abnormal heat loss event; extracting an abnormal event timestamp based on the abnormal heat loss event; and pressure stability evaluation is conducted in the corresponding time period according to the abnormal event timestamp, and when it is detected that the pressure change rate gradient is slowly decreased, it is judged that the heating system is in a water leakage state, and a water leakage water supplementing mode is started. The water leakage state of the heating system is accurately recognized, segmented intelligent water supplementing is achieved, and the operation stability and safety of the heating system are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of heating system water replenishment control, and in particular to an intelligent water replenishment control method and device for a heating system. BACKGROUND

[0002] During the long-term operation of the heating system, due to the aging of the internal pipeline, the deposition of scale, the wear of valve parts, and the degradation of sealing performance, the heat supply pipeline network is prone to problems such as leakage, micro-leakage, and even sudden leakage. At the same time, factors such as changes in user heating habits, system load fluctuations, and outdoor temperature changes can also cause large fluctuations in the supply and return water pressure and temperature, resulting in the destruction of the system heat balance. The above problems not only reduce heating efficiency and increase energy consumption, but also can cause frequent start-stop of the water pump, heat exchange equipment efficiency, and other hidden losses, and in severe cases, can even cause system circulation interruption, local pressure loss, or heating paralysis. In addition, if the water replenishment is not timely or the water replenishment amount is not accurate during the operation of the system, it will further exacerbate the pressure instability and cause a chain of failures. The traditional heating system water replenishment control method mainly relies on a fixed pressure threshold water replenishment controller or manual inspection judgment. When the system pressure is lower than the set value, the water replenishment is automatically started, and the water replenishment is stopped when the pressure rises. Although this method is simple in structure, it has obvious limitations: this control method is usually not sensitive to system leakage, slow leakage, and other hidden faults, and cannot provide effective early warning; the water replenishment behavior lacks fine management, often causing large pressure fluctuations. SUMMARY

[0003] To solve the above technical problems, the present application provides an intelligent water replenishment control method and device for a heating system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, the present application provides an intelligent water replenishment control method for a heating system, comprising the following steps: Step S1: calculating the supply and return water temperature difference value and the standardized enthalpy difference sequence of the water supply pipeline and the return water pipeline of the heating system; Step S2: collecting the rated configuration information currently set for the heating system, and calculating the theoretical temperature difference reference value according to the rated configuration information; Step S3: performing temperature difference deviation analysis on the theoretical temperature difference reference value according to the supply and return water temperature difference value and the standardized enthalpy difference sequence, and marking abnormal heat loss events; Step S4: extracting an abnormal event timestamp based on the abnormal heat loss event; performing pressure stability evaluation for the corresponding time period according to the abnormal event timestamp, and determining that the heating system is in a water leakage state when a gentle downward gradient of the pressure change rate is detected, and entering a water leakage replenishment mode; Step S5: calculating a water replenishment demand value based on the water leakage replenishment mode; performing segmented water replenishment control according to the water replenishment demand value, and completing the water replenishment operation.

[0005] In the present specification, an intelligent water replenishment control device of a heating system is provided for executing the intelligent water replenishment control method of the heating system as described above, comprising: a temperature difference calculation module for calculating a supply-return water temperature difference value of the supply pipeline and the return pipeline of the heating system and a standardized enthalpy difference sequence; a reference module for collecting the rated configuration information currently set by the heating system, and calculating a theoretical temperature difference reference value according to the rated configuration information; a deviation analysis module for performing temperature difference deviation analysis on the theoretical temperature difference reference value according to the supply-return water temperature difference value and the standardized enthalpy difference sequence, and marking an abnormal heat loss event; a pressure evaluation module for extracting an abnormal event timestamp based on the abnormal heat loss event; performing pressure stability evaluation for a corresponding time period according to the abnormal event timestamp; when a pressure change rate gradient is detected to be gently descending, it is determined that the heating system is in a water leakage state, and enters a water leakage replenishment mode; a water replenishment control module for calculating a water replenishment demand value based on the water leakage replenishment mode; performing segmented water replenishment control according to the water replenishment demand value, and completing water replenishment operation.

[0006] The beneficial effects of the present application are as follows: by calculating the temperature difference between the supply water and the return water in real time and the corresponding standardized enthalpy difference sequence, the actual heat exchange efficiency and heat loss level of the system can be more accurately represented, which can better reflect the thermal state of the system compared to using only temperature values. The standardized enthalpy difference sequence eliminates the influence of different working conditions on the heat parameters, making the subsequent comparison with the theoretical benchmark more reliable and improving the sensitivity and accuracy of anomaly detection. By collecting configuration parameters such as design supply and return water temperature, rated flow, and rated heat dissipation, and calculating the theoretical temperature difference, the judgment basis is highly matched with the specific system, avoiding the error of "one-size-fits-all" fixed threshold. By comparing the theoretical benchmark value with the actual data, false positives and false negatives can be significantly reduced, ensuring that the water replenishment logic is based on real working condition deviations. Whether it is central heating, individual heating, or variable flow systems, the corresponding benchmark can be generated based on the rated parameters, improving the universality of the scheme. When the temperature difference and enthalpy difference deviate from the theoretical benchmark, it can prompt the existence of additional heat loss, providing a pre-signal for subsequent judgment of leakage and other faults. Through sequence analysis, persistent deviations can be identified, avoiding being misled by short-term user heating demand changes, effectively improving the reliability of anomaly event detection. The marked abnormal heat loss events can form a time sequence, providing key data support for subsequent pressure gradient-based leakage trend analysis. Judging from temperature difference or pressure alone can easily lead to misjudgment, so this step combines the characteristics of both signals. Only when an abnormal heat loss event occurs and is accompanied by a gentle decline in pressure gradient is it determined to be a leak, significantly reducing the false negative and false positive rates. Compared to traditional threshold detection, this method can capture early and slow leakage trends, enabling early warning. The method achieves automated processing without human intervention, avoiding the risk of reduced system circulation efficiency or shutdown due to prolonged standby, while providing continuous and stable heating effects for users. According to the water replenishment demand calculated from the leakage, the problem of "over-replenishment" or "under-replenishment" in traditional water replenishment methods is avoided, thereby maintaining stable system pressure. The gradual and multi-level water replenishment method can reduce the risk of water hammer, equipment impact, and pipeline noise caused by instantaneous large-scale water replenishment, improving system safety. Stable pressure can maintain the normal operation of the circulating pump and heat exchange equipment, reducing the loss caused by frequent start-stop. The entire process from anomaly detection, state determination to water replenishment execution is automatically completed, improving the level of intelligence and enabling unattended operation. BRIEF DESCRIPTION OF DRAWINGS

[0007] Fig. 1 Figure 1 is a schematic diagram of the steps of the intelligent water replenishment control method of the heating system of the present application; Fig. 2 Figure 2 is a schematic diagram of the detailed implementation steps of step S1; Fig. 3 Figure 3 is a schematic diagram of the detailed implementation steps of step S2. DETAILED DESCRIPTION

[0008] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0009] This application provides an intelligent water replenishment control method and device for a heating system. The executing entities of the intelligent water replenishment control method and device for the heating system include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc., which can be considered as general computing nodes in this application. The data processing platform includes, but is not limited to, at least one of an audio-visual management system, an information management system, and a cloud data management system.

[0010] Please see Figs. 1 to 3 This invention provides an intelligent water replenishment control method for a heating system, comprising the following steps: Step S1: Calculate the temperature difference between the supply and return water in the heating system and the standardized enthalpy difference sequence; Step S2: Collect the rated configuration information of the heating system currently set, and calculate the theoretical temperature difference benchmark value based on the rated configuration information; Step S3: Perform temperature difference deviation analysis on the theoretical temperature difference benchmark value based on the supply and return water temperature difference value and the standardized enthalpy difference sequence, and mark abnormal heat loss events; Step S4: Extract the timestamp of the abnormal heat loss event; perform pressure stability assessment for the corresponding time period based on the timestamp of the abnormal event. When a gradual decrease in the pressure change rate gradient is detected, the heating system is determined to be in a leaking state and enters the leak replenishment mode. Step S5: Calculate the water replenishment demand value based on the water leakage replenishment mode; perform segmented water replenishment control according to the water replenishment demand value to complete the water replenishment operation.

[0011] In the embodiments of the present invention, see Fig. 1 The diagram below illustrates the steps of an intelligent water replenishment control method for a heating system according to the present invention. In this example, the steps of the intelligent water replenishment control method for the heating system include: Step S1: Calculate the temperature difference between the supply and return water in the heating system and the standardized enthalpy difference sequence; In this embodiment, an array of temperature sensors deployed on the supply and return water pipes of the heating system is used to collect temperature data in real time during operation. For example, the supply water temperature sensor T_s records the current supply water temperature with a sampling period of 1 second, and the normal range may be between 45℃ and 75℃; the return water temperature sensor T_r records the return water temperature simultaneously, which is usually between 35℃ and 65℃. The system performs basic validity checks on the collected temperature sequence, including checking whether the temperature jump is within the allowable range and whether there are abnormal peaks, and then calculates the supply and return water temperature difference ΔT = T_s - T_r. This temperature difference is the core parameter for measuring the effective heat transfer capacity of the system. If ΔT is consistently low, it usually indicates that heat is dissipated prematurely in the pipe network or that indoor heating is insufficient. Subsequently, the system combines the real-time flow rate Q(t) provided by the electromagnetic flowmeter and uses the conventional heat power calculation formula P(t) = c·ρ·Q(t)·ΔT(t) to fuse the temperature difference and flow rate to generate an instantaneous enthalpy difference sequence. Since the raw data is affected by flow fluctuations, pump frequency changes, or sensor noise, the system uses time-domain filtering methods, such as moving average or median filtering, to smooth the enthalpy difference sequence and obtain the standardized enthalpy difference sequence Z_h(t) through normalization.

[0012] Step S2: Collect the rated configuration information of the heating system currently set, and calculate the theoretical temperature difference benchmark value based on the rated configuration information; In this embodiment, the currently set rated operating parameters are read from the heating control system. These typically include the boiler's rated output power (e.g., 150 kW), the designed supply and return water temperatures (e.g., set supply water temperature 70℃, set return water temperature 50℃), and the rated circulation flow rate (e.g., 9 m³ / h). 3 / h). These rated parameters represent the design heat transfer capacity and temperature difference requirements that the system should achieve under normal operating conditions. Subsequently, the system calculates the unit heat load requirement by combining the building's design thermal performance (such as wall heat transfer coefficient, insulation strength, etc.) and the outdoor ambient temperature, and derives the theoretical supply and return water temperature difference through a heat balance model. For example, when the outdoor temperature is -3℃ and the heat loss coefficient of the building envelope is 65 W / m². 2 The system calculates the instantaneous heat demand as a specific value based on the heat load model, and calculates the theoretical temperature difference ΔT_theory = P_load / (c·ρ·Q_rated) based on the relationship between flow rate and heat power. The calculated theoretical temperature difference value is usually in the range of 10 to 20℃, serving as the benchmark temperature difference for the heating system under ideal conditions. This benchmark is used to compare the actual supply and return water temperature difference with the enthalpy difference to determine whether there is any abnormal heat loss in the system. In this way, the theoretical temperature difference benchmark value becomes a key reference indicator in subsequent deviation analysis.

[0013] Step S3: Perform temperature difference deviation analysis on the theoretical temperature difference benchmark value based on the supply and return water temperature difference value and the standardized enthalpy difference sequence, and mark abnormal heat loss events; In this embodiment, after obtaining the actual supply and return water temperature difference ΔT_real and the theoretical temperature difference benchmark ΔT_theory, the system performs temperature difference deviation analysis. Deviation is typically calculated using D(t) = ΔT_theory - ΔT_real. If the deviation value remains consistently large in the short term, for example, exceeding a preset threshold (e.g., 5°C or 30% of the design value), it indicates that the system's heat is not being effectively transferred to the end, and may be prematurely consumed by factors such as pipe leaks, external heat dissipation, or abnormal water mixing. Simultaneously, the system references the trend of the standardized enthalpy difference sequence Z_h(t). If the enthalpy difference trend is inconsistent with the theoretical heat transfer trend, such as a rapid decline in enthalpy difference or a prolonged period below the standard value, it further indicates that the system may be experiencing abnormal heat loss. By synchronously comparing the ΔT deviation with Z_h(t), the system can extract abnormal feature points, such as the duration of the deviation and the slope of the deviation. Finally, if the judgment logic conditions are met (such as the deviation lasting for more than 8 seconds and the second derivative of the enthalpy difference sequence being negative), the moment is marked as an abnormal heat loss event and the timestamp is recorded.

[0014] Step S4: Extract the timestamp of the abnormal heat loss event; perform pressure stability assessment for the corresponding time period based on the timestamp of the abnormal event. When a gradual decrease in the pressure change rate gradient is detected, the heating system is determined to be in a leaking state and enters the leak replenishment mode. In this embodiment, after marking the abnormal heat loss event, the system extracts the corresponding timestamp t_abn as a key node, and then retrieves the expansion tank pressure data P(t) within a certain time range before and after the abnormality (e.g., 20 seconds before and 20 seconds after). The system performs filtering and baseline correction on the pressure sequence to remove random sensor disturbances, and calculates the pressure change rate dP / dt and the second-order change rate d 2 P / dt 2 If, near the abnormal moment, the pressure change rate remains negative or exhibits a slow decreasing trend, for example, gradually decreasing from 0.230 MPa to 0.224 MPa, with the rate of decrease remaining stable at around -0.0005 MPa / s and without significant rebound, then it can be determined that the system exhibits stable leakage behavior. Simultaneously, if d 2 P / dt 2 The pressure readings also remained in the negative range, indicating that the pressure drop was not a sudden shock but a continuous depressurization process, further enhancing the accuracy of the leak detection. After comprehensively analyzing the above pressure indicators, the system ultimately determined that the system was experiencing liquid loss and had entered a leaking state. At this point, the controller would switch to the leak-replenishment mode and suspend other non-essential adjustment processes to ensure that the replenishment logic would be executed first.

[0015] Step S5: Calculate the water replenishment demand value based on the water leakage replenishment mode; perform segmented water replenishment control according to the water replenishment demand value to complete the water replenishment operation.

[0016] In this embodiment, after entering the water leakage replenishment mode, the system calculates the pressure loss ΔP_loss based on the difference between the current pressure value of the expansion tank and the normal pressure reference value (e.g., 0.25 MPa), and converts the missing pressure into the corresponding water shortage volume according to the system hydraulic model. For example, when ΔP_loss is 0.02 MPa, based on the effective volume of the expansion tank of 40 L and the corresponding pressure curve, the water shortage can be calculated to be approximately 2-3 L. Simultaneously, the system combines the heat loss reflected by the enthalpy difference sequence to adjust the water replenishment demand, avoiding insufficient water replenishment due to relying solely on pressure. After calculating the final water replenishment demand, the water replenishment execution system adopts a segmented water replenishment strategy, dividing the water replenishment into an initial small-flow water replenishment and a subsequent accelerated water replenishment to ensure safety and efficiency. For example, the first segment performs initial water replenishment at 0.4 L / min to steadily increase the pressure; the second segment performs water replenishment at 1.0 L / min to smoothly increase the system pressure to the reference range. During the water replenishment process, the system continuously monitors the pressure trend and gradually reduces the water replenishment rate when the target pressure value (e.g., 0.245–0.250 MPa) is reached, eventually closing the water replenishment valve to complete the replenishment operation. The entire process ensures safe and reliable pressure recovery after leakage and prevents pressure surges or system instability caused by excessively rapid water replenishment.

[0017] In this embodiment, see Fig. 2 The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Real-time temperature parameters of the heating system's supply and return water pipes are collected using an online sensor array. Calculate the temperature difference of the real-time temperature parameters to obtain the supply and return water temperature difference; Real-time water flow data of the water supply and return pipelines are collected using electromagnetic flowmeters to obtain time-series flow curves. Instantaneous heat power transfer is calculated based on the time-series flow rate curve and the real-time temperature parameters to generate a supply and return water enthalpy difference sequence. The time-domain noise filter was applied to the enthalpy difference sequence of the supply and return water to obtain a standardized enthalpy difference sequence.

[0018] In this embodiment, temperature sensor arrays are deployed at key nodes of the pipeline network. Platinum resistance thermometers (such as Pt100 and Pt1000) or high-temperature thermocouples (such as type K) are commonly used, with accuracies typically ranging from ±0.1℃ to ±0.3℃ and a rapid response time of 1 to 3 seconds. The sensors are fixedly installed in the main pipeline or bypass section using a metal sleeve insertion method. The bypass method reduces the impact of mainstream flow on the measurement while ensuring the sensor can stably conform to the water flow to obtain accurate temperatures. In actual engineering deployments, the array typically includes multiple installation points such as the main water supply section, heat source outlet section, heat exchange branch nodes, and the return water main section, improving the reliability of temperature field judgment through multi-point monitoring. Each sensor outputs data to a PLC or industrial control computer in real time via 4–20mA or RS485 (Modbus RTU). The sampling period can be set to 1 second or 5 seconds to adapt to the dynamic change rate of different systems. In the experimental conditions, the typical supply water temperature varies between 50 and 70°C, while the return water temperature is usually maintained between 35 and 55°C. Therefore, the data acquisition process needs to be configured with a temperature compensation function to automatically correct the zero-point drift of the sensor caused by changes in ambient temperature. To extract the thermal change information between the supply and return water from the temperature monitoring data, the supply water temperature Ts and the return water temperature Tr acquired in step one need to be calculated in real time to obtain the supply and return water temperature difference ΔT. Although the calculation formula ΔT=Ts-Tr seems simple, in actual engineering implementation, the host computer system will perform time synchronization, anomaly detection, and smoothing of the temperature data. Since the supply and return water sensors may be located in different pipe sections, there may be communication delays ranging from hundreds of milliseconds to several seconds. Therefore, the program needs to align the temperature values ​​based on timestamps, and if necessary, use linear interpolation or sample expansion to ensure that the two temperatures correspond to the same measurement time. To reduce temperature difference fluctuations caused by noise, the system typically uses a 5-point moving average, low-pass smoothing, or Kalman filtering method to preprocess the raw temperature, making the ΔT change closer to the actual thermal conditions. In experimental conditions, such as a supply water temperature maintained at 60℃ and a return water temperature of approximately 45℃, the calculated temperature difference is approximately 15℃. However, in actual operation, due to changes in valve opening, fluctuations in user-side heat dissipation load, or adjustments in the circulation pump frequency, the temperature difference may exhibit periodic or instantaneous disturbances. Instantaneous flow rate is obtained by installing an electromagnetic flowmeter. The electromagnetic flowmeter utilizes Faraday's law of electromagnetic induction to generate a magnetic field within the pipe and measures the potential difference generated by the fluid cutting magnetic field lines, which is then converted into flow velocity and volumetric flow rate. To ensure measurement accuracy, the flowmeter should be installed with straight pipe sections of at least 5D and 3D (D being the pipe diameter) before and after it, respectively, to avoid interference from eddy currents caused by valves, elbows, or tees. The typical accuracy of the flowmeter is ±0.5%, the sampling period can be set to 0.5 seconds or 1 second, and it transmits data to the control system via 4–20mA or RS485 communication.Since the supply and return water flow rates should be essentially the same in a closed heating system, the flow time-series curve not only reflects the system's circulation capacity but can also identify faults such as leaks, blockages, or pump performance degradation through abnormal deviations. In experimental conditions, for example, the instantaneous supply water flow rate is approximately 8 m³ / s. 3 / h, while the return water flow rate is approximately 7.9 m³ / h. 3 Slight differences in flow rate per hour ( / h) may arise from local branch throttling or measurement errors. During the data processing phase, to obtain a smooth and reliable flow time series curve, the system stores the continuously sampled flow sequences in chronological order and uses median filtering or low-frequency smoothing algorithms to remove spike noise.

[0019] After obtaining real-time sequence data of flow rate and temperature difference, it is necessary to dynamically calculate the instantaneous heat power of the system to obtain the enthalpy difference sequence reflecting the change in heat transport capacity. Based on hydraulic-thermal theory, the formula for calculating instantaneous heat power P(t) is: P(t) = ρ × Cp × Q(t) × ΔT(t), where ρ is the density of water (generally ranging from 997 to 1000 kg / m³, varying with temperature). 3Cp represents the specific heat capacity of water (approximately 4.186 kJ / (kg·℃)), and Q(t) and ΔT(t) are the time-series flow rate and temperature difference obtained in steps two and three, respectively. To ensure calculation accuracy, the system typically performs automatic corrections to ρ and Cp based on the supply and return water temperatures. For example, when the system temperature rises to around 70℃, the density of water decreases by approximately 2%, and failure to correct this could lead to accumulated errors in thermal power. The calculation process requires all data to be synchronized with a uniform sampling period. Therefore, the program matches flow rate and temperature difference values ​​according to timestamps. If data on one side is missing, an interpolation completion strategy is used to avoid calculation interruption. Under experimental conditions, such as an instantaneous flow rate Q(t) = 2.0 kg / s and a temperature difference ΔT(t) = 12℃, the instantaneous thermal power is approximately 100 kW. Continuously calculating each sampling point yields the enthalpy difference sequence P(t) that varies with time. This sequence clearly reflects characteristics such as changes in user heat load, system thermal balance, and pump frequency. Enthalpy difference sequence is also an important basis for judging whether the system has a decrease in heat exchange capacity, radiator blockage, or insufficient heating. Since the enthalpy difference sequence is calculated from data from multiple real-time sensors, it inevitably contains high-frequency noise, spike interference, and measurement errors. Therefore, it must be filtered and standardized in the time domain before being used for intelligent water replenishment control. Noise characteristic analysis reveals that the disturbances in the enthalpy difference sequence mainly include sensor jitter, sudden flow changes, valve action interference, and external load disturbances. Therefore, various filtering techniques can be used, including moving average filtering, median filtering, first-order low-pass filtering, or dynamic filters based on Kalman estimation. Among them, median filtering is suitable for removing occasional spikes, while low-pass filtering can effectively suppress high-frequency noise, allowing the enthalpy difference curve to reflect the true trend of thermal change. In experimental scenarios, the original enthalpy difference sequence often exhibits fluctuations of ±5% to ±8%, which can be significantly converged to ±1% to ±2% after filtering, facilitating stable threshold judgment and algorithm modeling. After filtering, the sequence needs to be standardized. For example, Min-Max normalization can be used to scale the data to the 0-1 range, or Z-score normalization can be used to give it the statistical characteristics of a mean of 0 and a standard deviation of 1. The standardized enthalpy difference sequence has higher contrast, stability, and computability, and can be directly used in intelligent water replenishment decision models, such as determining whether the system's heat loss exceeds a threshold and inferring the trend of changes in water replenishment demand.

[0020] In this embodiment, see Fig. 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Collect the current rated configuration information of the heating system, including the boiler rated power, design supply and return water temperature and rated circulation flow rate; Obtain outdoor ambient temperature and thermal parameters of the building envelope; Transient heat load is calculated based on the outdoor ambient temperature and the thermal parameters of the building envelope to obtain the instantaneous heat load demand; The theoretical supply and return water temperature difference is predicted based on the rated configuration information and instantaneous heat load demand to obtain the theoretical temperature difference benchmark value.

[0021] In this embodiment, in the heating system, rated configuration information is the fundamental parameter for determining system operating capacity, estimating theoretical operating conditions, and establishing a reference model for water replenishment control. This step requires obtaining the rated thermal power (e.g., 1.5MW, 2MW, 5MW, etc.) of the boiler or heat exchanger unit from system design documents, unit nameplates, or the system control platform, and reading the design supply and return water temperatures, such as typical 75 / 55℃, 70 / 50℃, or 60 / 40℃ operating conditions. These temperature differences reflect the system's heat transfer capacity under design conditions. Simultaneously, design circulation flow parameters also need to be collected, for example, a system rated flow of 40 m³ / h. 3 / h、60 m 3 / h or 100m 3 The flow rate, measured in h, directly affects the supply and return water temperature difference, the unit's heat transfer coefficient, and the hydraulic balance of the entire system. In practical engineering implementation, these parameters are typically collected through two methods: firstly, by reading digital configuration parameters from the BAS (Building Automation System) or heat source control system; and secondly, by manually entering the rated values ​​from the system design specifications. To ensure data accuracy, the system needs to verify the consistency of these rated parameters with actual operating parameters. For example, checking the boiler nameplate confirms whether the boiler's rated power is 2MW, and monitoring the current operating curve confirms whether the system is operating at the designed flow rate. To predict the building's current real-time heat load demand, a thorough understanding of external meteorological conditions and the building's own thermal characteristics is necessary. The system acquires the outdoor ambient temperature in real time through temperature sensors installed on the building's exterior walls, weather station data interfaces, or urban meteorological service platforms. For example, the current outdoor temperature may be -8℃, 2℃, or 10℃, and different outdoor temperatures will cause significant changes in the building's heat load. Simultaneously, it is also necessary to obtain the thermal performance indicators of the building envelope, including the external wall heat transfer coefficient K-value (e.g., 0.45 W / m²). 2 ·K, 0.6 W / m 2 • K), window heat transfer coefficient (e.g., 2.0 W / m²) 2 Parameters such as heat transfer coefficient (K), roof heat transfer coefficient, building shape coefficient, and infiltration ventilation rate are typically obtained from building design and construction data or energy-saving assessment documents. In some systems, where complete design data is lacking, infrared thermal imaging, thermal inertia estimation, or simplified models can be used to infer the heat transfer parameters of the building envelope. After collecting these parameters, the system will provide a structured description based on the building area, exposed area of ​​the building envelope, and material classification. For example, for an exterior wall area of ​​1200 m²... 2 Window area 300 m² 2Roof area 600 m 2 Each corresponds to a different heat transfer coefficient and thermal stability performance.

[0022] After obtaining the building envelope parameters and outdoor temperature, the transient heat load demand of the building needs to be obtained through thermal calculations. Heat load calculation typically uses the building heat transfer formula: Q = K × A × (Tin - Tout), where K is the overall heat transfer coefficient of the building envelope, A is the heat exchange area of ​​the building envelope, Tin is the indoor set temperature, and Tout is the outdoor ambient temperature. To obtain a more realistic transient heat load, the system further incorporates calculations of heat loss due to air infiltration, such as Qv = 0.278 × V × n × (Tin - Tout), where V is the building's internal volume and n is the number of air infiltration cycles. In actual engineering, these calculations can be performed automatically using a heat load model module, updating the heat load results in real time based on changes in outdoor temperature and building envelope performance. For example, in experimental conditions, the building area is 5000 m². 2 The average heat transfer coefficient of the building envelope is 0.6 W / m. 2 ·K, with an indoor design temperature of 20℃, when the outdoor temperature is -5℃, the heat load required for heat transfer from the building envelope alone is: Q≈0.6×5000×(20-(-5))=75kW. Adding the heat loss from infiltration ventilation of approximately 25kW, the total heat load is approximately 100kW. The heat load may increase further when the outdoor temperature decreases or the wind speed increases. The system calculates the above model in real-time every 1 minute or 5 minutes to generate a dynamically changing instantaneous heat load demand curve. After obtaining the system's rated configuration information (rated power, design supply and return water temperatures, rated flow rate) and instantaneous heat load demand, a theoretical model needs to be constructed to predict the supply and return water temperature difference that the system should achieve at this time, providing a benchmark reference for intelligent water replenishment control. Theoretical prediction usually follows the heat conservation formula: ΔT_theory = Q_load / (ρ×Cp×Qdesign), where Q_load is the instantaneous heat load obtained in step three, Q_design is the rated design circulation flow rate, and ρ and Cp are the density and specific heat capacity of water, respectively. The model combines the building's actual heat demand with system design parameters to calculate the supply and return water temperature difference that the system should maintain under ideal operating conditions. For example, in experimental conditions, if the instantaneous heat load is 120 kW and the design circulation flow rate is 60 m³ / h... 3If the load is 16.7 kg / s (equivalent to 120000 / (4186 × 16.7) ≈ 1.7℃), then the theoretical temperature difference is approximately ΔT_theory = 120000 / (4186 × 16.7) ≈ 1.7℃. If the load increases to 300 kW, the theoretical temperature difference will rise to approximately 4.5℃. During the calculation process, the system also incorporates constraints such as boiler rated power limits and pump capacity limits. For example, when the instantaneous heat load exceeds the rated power, the theoretical temperature difference will be calculated based on the maximum capacity and will not increase indefinitely; automatic correction will also be performed when the flow rate deviates from the rated value.

[0023] In this embodiment, step S3 includes the following steps: The supply and return water temperature difference is compared with the theoretical temperature difference benchmark value in real time to calculate the instantaneous temperature difference deviation. The instantaneous temperature difference deviation is fitted over time to obtain a deviation parameter set. Based on a preset supply and return water deviation threshold, a sliding window statistical analysis is performed on the deviation parameter set to mark the time points that exceed the supply and return water deviation threshold. Based on the time points, the standardized enthalpy difference sequence is synchronously matched to extract the enthalpy difference information at the corresponding time points; Calculate the rate of change of enthalpy difference and the second derivative based on enthalpy difference information; Trend fluctuation analysis is performed based on the rate of change of enthalpy difference and the second derivative. When the trend fluctuation of enthalpy difference is abnormal, it is determined to be an abnormal heat loss event.

[0024] In this embodiment, to determine in real time whether the heating efficiency of the heating system deviates from normal operating conditions, it is necessary to compare the actual measured supply and return water temperature difference with the theoretically predicted temperature difference benchmark to obtain the instantaneous temperature difference deviation. The actual temperature difference is collected by a temperature sensor array, while the theoretical temperature difference is calculated from the system's rated configuration parameters and the building's instantaneous heat load. During system operation, the supply and return water temperature difference is collected every second or every 5 seconds, and it is timestamped with the theoretical temperature difference at the same time to ensure that the two temperature difference signals correspond to the same thermal state at the same moment. For example, under experimental conditions, if the actual temperature difference is 8.5℃ and the theoretical temperature difference is 6.0℃, the instantaneous temperature difference deviation is 2.5℃; if the actual temperature difference is lower than the theoretical value, such as 4.0℃ for the actual difference and 6.0℃ for the theoretical difference, the deviation is -2.0℃. The sign of the deviation reflects different operating states. A positive deviation may indicate excessive flow or overheating by the system, while a negative deviation usually means insufficient heat exchange, water shortage, or increased heat loss. To avoid judgment errors caused by instantaneous fluctuations, the system uses moving average or exponentially weighted smoothing methods to smooth both the actual and theoretical temperature differences, making the deviation more stable and trend-oriented. Simultaneously, a dynamic limiter prevents extreme values ​​(such as sudden temperature jumps caused by sensor malfunctions) from interfering with the calculation results. Instantaneous temperature difference deviations typically fluctuate dynamically with operating conditions, load changes, and pump frequency variations. Therefore, time-series fitting and statistical analysis of the deviation data are necessary to identify persistent or sudden deviations. This step involves time-series fitting of the deviation time series, which can employ linear fitting, polynomial fitting, or exponential smoothing methods to construct a fitting curve that better characterizes the system's trajectory based on data trends. For example, in experimental conditions with a sampling frequency of 1Hz, a 60-second time interval can be used to perform first- or second-order polynomial fitting on the deviation series, extracting parameters such as the deviation change trend, growth rate, and curve slope to form a deviation parameter set. Subsequently, to identify whether the deviation has reached an abnormal level, the system performs sliding window statistical analysis on the fitted deviation based on a preset supply and return water temperature difference deviation threshold (e.g., ±2℃ or ±3℃). The window size can be set to 30 seconds, 60 seconds, or 120 seconds to capture the persistence of the deviation over a short period of time, rather than an instantaneous peak. Within the sliding window, the number, percentage, and duration of sample points with deviations exceeding the threshold are statistically analyzed. For example, if a time deviation exceeds the threshold for more than 25 seconds within a 60-second window, the corresponding time period can be marked as an abnormal deviation time point.

[0025] After obtaining the time points of deviation anomalies, these time points need to be synchronized and matched with the standardized enthalpy difference sequence to extract the corresponding enthalpy difference signals for further analysis. The enthalpy difference sequence is calculated from multi-dimensional signals such as flow rate and temperature difference, and after previous filtering and standardization, it has the characteristics of low noise and clear trend. To ensure data synchronization, the system needs to align the enthalpy difference sequence and the deviation sequence using a unified time base (such as a Unix timestamp), and find the enthalpy difference value corresponding to each deviation anomaly time point through integer second matching or linear interpolation. For example, in the experimental conditions, the deviation anomaly time points are t=300s, 540s, and 860s, while the enthalpy difference sequence sampling frequency is once per second, so the enthalpy differences H(300), H(540), and H(860) can be directly extracted; if the sampling frequencies are inconsistent, such as deviation data 1Hz and enthalpy difference data 0.5Hz, then the equivalent enthalpy difference value at t=300s needs to be interpolated. To improve reliability, the system extends the time window forward and backward by a certain amount (e.g., ±10 seconds or ±30 seconds) from the abnormal time point to extract the corresponding local enthalpy difference segments, observing the evolution of enthalpy difference changes before and after the anomaly. Changes in enthalpy difference reflect not only heat power transfer capacity but also changes in the system's hydraulic state, such as an increase in enthalpy difference due to insufficient water replenishment or a decrease due to pipeline leakage. After obtaining the enthalpy difference segments corresponding to the abnormal time points, further mathematical analysis of the enthalpy difference's characteristics over time is required. This step quantitatively describes the rate and trend of enthalpy difference change by calculating the first derivative (rate of change) and second derivative (curvature) of the enthalpy difference. The rate of change of enthalpy difference ΔH / Δt can be obtained through the difference method between adjacent sampling points; for example, with a 1-second sampling interval, the rate of change is equal to H(t) - H(t-1). Under experimental conditions, if the enthalpy difference increases from 100kW to 110kW within 5 seconds, the rate of change is approximately 2kW / s, indicating a rapid change in heat load or flow rate during that period. The second derivative further reflects the rate of change, i.e., the acceleration of the enthalpy difference change. For example, the second derivative is (ΔH / Δt(t) - ΔH / Δt(t-1)) / Δt, which can be used to identify inflection points or abrupt changes. If the first derivative continues to increase, it means that the enthalpy difference is accelerating, which may correspond to a rapid increase in heat loss; if the second derivative is negative and has a large absolute value, it indicates that the system's heat exchange capacity may be rapidly decreasing. To avoid abnormal derivative values ​​caused by noise, the system calculates the derivative using methods such as smoothing differences, weighted slopes, or local polynomial fitting to make the results more stable. For example, using 5-point difference fitting to obtain the first derivative and 7-point local fitting to obtain the second derivative can significantly improve the stability of the derivative. After obtaining the rate of change of enthalpy and its second derivative, trend fluctuation analysis is required to determine whether there are any abnormal heat loss events in the system. The core of trend analysis lies in identifying whether the enthalpy curve exhibits abnormal, rapid, accelerated, or periodic disturbances.For example, if the first derivative remains at a high value for a short period of time (e.g., greater than 1.5 kW / s) and the duration exceeds a set threshold (e.g., 10 seconds), it indicates that there are signs of rapid changes in the system's heat load or hydraulic state; if at the same time the second derivative shows a continuous extreme value, for example, continuously positive and greater than 0.5 kW / s. 2 If the enthalpy difference is increasing rapidly, it may correspond to increased local heat loss in the pipeline network or insufficient water supply leading to a decrease in flow rate. Conversely, if the enthalpy difference is continuously decreasing and the second derivative is negative, it may indicate a decrease in heating capacity, insufficient boiler power, or abnormal operation of the circulating pump. Under experimental conditions, if the enthalpy difference decreases from 110kW to 90kW within 20 seconds during a certain event period, with an average rate of change of -1.0 kW / s and the second derivative exhibits continuous negative fluctuations, then an abnormal heat loss trend can be identified during that period. The system sets logical detection conditions for these trend characteristics, such as combinations of conditions like "the rate of change exceeds the threshold and the duration meets the requirements" and "the second derivative and the rate of change both point in the abnormal direction." When multiple conditions are met simultaneously, an abnormal event can be identified.

[0026] In this embodiment, step S4 includes the following steps: Real-time calculation of pressure monitoring data in the expansion tank; The pressure monitoring data is digitally filtered to obtain optimized pressure data; Extracting timestamps of abnormal events based on abnormal heat loss events; The pressure data is time-matched based on the timestamps of abnormal events, and the pressure change rate within that time period is calculated. Pressure stability is assessed based on the rate of pressure change. When a gradual decrease in the rate of pressure change is detected, the heating system is determined to be in a leaking state and enters the leak replenishment mode.

[0027] In this embodiment, a digital pressure sensor, such as a diaphragm pressure transmitter or a strain gauge pressure transmitter, is typically installed on the side wall or top of the expansion tank. The sensor's range is generally 0–0.6 MPa or 0–1.0 MPa, with an accuracy of 0.25%–0.5%FS. Pressure values ​​can be acquired via a 4–20 mA current signal or RS485 (Modbus) output. The system reads the pressure signal at a fixed sampling period of 1 or 2 seconds, converting the input electrical signal into actual water pressure through a linear mapping relationship; for example, 4 mA corresponds to 0 MPa, and 20 mA corresponds to the upper limit of the device's range. Real-time pressure data is continuously recorded to form a pressure sequence, and communication error values, out-of-range signals, or sudden jumps are eliminated through primary data verification rules, such as a pressure value that momentarily deviates significantly from the normal range (e.g., a system with a pressure of 0.45 MPa suddenly reaching 1.5 MPa). To construct higher-precision pressure baseline data, the system also simultaneously records auxiliary information such as timestamps, water supply valve status, and circulating pump operating status. Because pressure sensor data often includes equipment noise, pipeline pulsation, pump frequency fluctuations, or instantaneous spikes caused by valve opening and closing, the raw pressure monitoring data needs to be digitally filtered to obtain a stable, continuous pressure curve that accurately reflects the system trend. Filtering methods can include moving average filtering, weighted moving average, median filtering, or digital low-pass filtering, depending on the characteristics of the pressure fluctuations. For example, if the pressure signal exhibits high-frequency irregular vibrations, low-pass filtering can be used to suppress high-frequency interference; if occasional pressure spikes occur, median filtering can effectively eliminate sudden anomalies; if it is necessary to enhance the tracking of slow pressure trends, exponentially weighted moving averages can be used, giving higher weight to recent data and lower weight to older data. After filtering, the pressure curve will show a smoother trend, effectively reflecting the true pressure changes inside the expansion tank. For example, if there are slight fluctuations in system pressure from 0.230 MPa to 0.228 MPa, after filtering, it can be displayed as a smooth result of 0.229 MPa, eliminating unnecessary noise.

[0028] To establish a correlation between pressure behavior and the thermodynamic state of the heating system, it is necessary to obtain the specific time points when abnormal heat losses occur in the system. This step utilizes the abnormal heat loss event records obtained from previous analyses to extract the occurrence time of each anomaly. Abnormal heat loss events are typically identified when there are abnormal fluctuations in the enthalpy difference trend of the system's supply and return water, a decrease in heat transfer efficiency, or a temperature difference deviating from the model's predicted value beyond a threshold. Each abnormal event corresponds to an accurate timestamp, and the system organizes these timestamps into a time series for subsequent analysis. During the extraction process, the system further verifies the validity of the timestamps, for example, by comparing them with boiler start-up and shutdown, circulating pump status changes, and valve control actions in the heating system's operation log to exclude heat transients caused by normal equipment switching, thereby ensuring that the extracted time points are truly related to abnormal thermal events. If multiple events occur too close together, such as less than 30 seconds apart, they can be considered as the same abnormal process and merged to avoid repeatedly triggering pressure matching calculations. These timestamps will serve as key time-based references for subsequent pressure sequence alignment, pressure change rate calculation, and leakage trend judgment, enabling the system to focus on checking water pressure changes around these critical moments and accurately establish the correspondence between heat loss events and pressure anomalies. After identifying the timestamps of abnormal events, this step aligns these time points with the optimized pressure data sequence using time matching to analyze whether there are significant pressure changes before and after the abnormal event. Time matching is usually based on a unified time axis. For example, if the pressure data is sampled at 1-second intervals and the abnormal event time is accurate to the second, the corresponding points can be directly extracted from the pressure sequence. If the pressure data sampling frequencies are different, the pressure value at the precise corresponding moment can be calculated using linear interpolation. To more accurately assess the pressure change trend, the system not only extracts the pressure value at the moment of the event but also extracts a specific time window before and after the event, such as 30 seconds or 60 seconds before and after, forming a local pressure data segment. The pressure change rate is then calculated within this time period. The pressure change rate is usually calculated using a difference method, i.e., ΔP / Δt = P(t) - P(t-Δt), where Δt is determined according to the pressure sampling period. If the pressure shows a continuous downward trend over a certain period of time, for example, gradually decreasing from 0.230 MPa to 0.220 MPa within 120 seconds, the rate of change is approximately -0.000083 MPa / s, indicating that the system may be experiencing a stable pressure relief phenomenon due to insufficient water replenishment or leakage. If the rate of change fluctuates periodically, it may be related to changes in pump frequency or the operation of regulating valves.

[0029] After obtaining the pressure change rate sequence, this step analyzes the direction, amplitude, and trend of the pressure change rate to determine if the system is leaking. If the pressure change rate is consistently negative and exhibits a slow, stable decline pattern with a small amplitude, such as remaining between -0.00002 and -0.00008 MPa / s for an extended period, it indicates that the pressure inside the expansion tank is decreasing in a constant but gradual manner. This often corresponds to a hidden leak or insufficient water replenishment in the heating system. If the rate of change gradient shows a stable characteristic, i.e., the absolute value of the rate of change does not suddenly increase, and the slope of the curve is nearly constant but the direction is negative, it can be considered that the system has a stable pressure relief trend. Conversely, if the pressure change rate fluctuates significantly or shows a rapid decline, it is usually related to factors such as equipment startup or valve operation, and does not necessarily indicate a leak. After confirming a slow downward pressure trend through trend fitting methods, such as linear trend slope analysis and directional judgment after multi-point curve smoothing, the system will trigger a leak detection based on a predetermined threshold. When a leak is detected, the system automatically enters the leak replenishment mode, which includes opening the automatic water replenishment valve, increasing the pressure sampling frequency, limiting pump operation fluctuations, and maintaining the system pressure within the set range, thereby ensuring that the heating system maintains stable operation and avoiding insufficient heating capacity or equipment failure due to continuous pressure drop.

[0030] In this embodiment, the specific steps of step S5 are as follows: Based on the water leakage replenishment mode, the difference between the pressure monitoring data and the preset normal pressure reference value is calculated to obtain the water tank pressure loss value; The total heat loss of the system is calculated based on the supply and return water temperature difference and the standardized enthalpy difference sequence. The actual water shortage is calculated by back-calculating the missing water tank pressure value and the total heat loss of the system. The actual water shortage is corrected by pipeline expansion compensation to obtain the water replenishment demand value; Segmented water replenishment control is performed based on the water replenishment demand value to complete the water replenishment operation.

[0031] In this embodiment, the optimized pressure monitoring data collected in real time is compared with the normal pressure reference value preset by the system. The normal pressure reference value can be determined according to the design conditions of the heating system, for example, the initial operating pressure is set to 0.25 MPa, the maximum pressure limit is 0.35 MPa, and the minimum stable pressure limit is 0.18 MPa. When the system is in a normal closed state, the pressure in the expansion tank should fluctuate within this steady-state range. After entering the leakage replenishment mode, the system continuously compares the real-time pressure with this reference value and calculates the difference between the two. The difference is the tank pressure loss value, representing the pressure drop caused by the reduction in water volume. The pressure loss value can be defined as ΔP = Pref - Pcurrent, where Pref is the normal pressure reference value and Pcurrent is the real-time monitoring pressure. For example, if the normal reference pressure is 0.25 MPa, and the current monitoring pressure is 0.21 MPa, then the pressure loss value is 0.04 MPa. The system not only takes the instantaneous pressure loss value but also performs a weighted analysis of the pressure loss trend over a period of time, such as averaging the pressure loss over the past 1 or 2 minutes to reduce the impact of instantaneous fluctuations. To accurately assess the degree of energy loss due to leakage, the total heat loss of the system is calculated by combining the supply and return water temperature difference with a time-domain filtered standardized enthalpy difference sequence. The supply and return water temperature difference reflects the overall operating status of the heat exchange system. When leakage occurs, in addition to pressure drop, heat will also exhibit an abnormal loss trend in the pipe network. For example, if the supply water temperature remains at 65°C while the return water temperature gradually decreases from 55°C to 50°C, it indicates an increase in system heat loss. The standardized enthalpy difference sequence reflects the actual heat change per unit time between the supply and return water; after filtering, it can more accurately reflect the actual heat transfer changes. The system calculates the heat loss per unit time by integrating the temperature difference sequence and the enthalpy difference sequence, and then sums them over the corresponding time periods to obtain the total heat loss. Heat loss is usually expressed in kJ or kWh. For example, during a water leak, if the enthalpy difference sequence shows a persistently low value and the temperature difference remains below the theoretical temperature difference benchmark, the system will determine that there is continuous heat loss and calculate the cumulative heat loss accordingly.

[0032] Based on the structural parameters of the expansion tank, such as its effective volume, expansion coefficient, and pressure change curve corresponding to water temperature, pressure changes are correlated with actual water volume changes. Pressure loss can be converted into water loss volume in the expansion tank. For example, if a 0.01 MPa pressure change in the expansion tank corresponds to a 0.8 L water volume change, then a pressure loss of 0.04 MPa corresponds to a water loss of approximately 3.2 L. However, relying solely on pressure conversion may be affected by temperature fluctuations and changes in the system's expansion coefficient; therefore, it needs to be corrected by incorporating heat loss data. Heat loss can be used to estimate the leaked water volume: for example, if the system has a supply water temperature of 65°C and a return water temperature of 50°C, and a total heat loss of 2.5 kWh is detected, then the corresponding water loss is Q = m·Cp·ΔT. That is, the water mass can be calculated using the specific heat capacity of water and the temperature difference, and then converted into volume. The system weightedly combines the water volume estimated from pressure and the water volume estimated from heat to obtain a more accurate actual water shortage. The fusion method employs a weighted allocation approach, such as a weight of 0.6 for the pressure method and 0.4 for the heat method, to offset the bias of a single method. Since the heating system's pipe network expands and contracts under different temperature conditions, after obtaining the actual water shortage, a pipe network expansion compensation correction is necessary to arrive at a more accurate replenishment demand. The compensation calculation is based on factors such as pipe network material, pipe diameter and length, and average water temperature. For example, common steel pipes can experience a volume change of 0.01% to 0.03% when the temperature changes by 10°C. If the system's pipe network volume is 800 L, the temperature change may lead to a volume expansion or contraction of approximately 0.08 to 0.24 L. The system calculates the expansion compensation based on the deviation between the real-time monitored average return water temperature and the set steady-state temperature, and deducts this from the actual water shortage. If the expansion tank itself expands due to temperature changes, corresponding compensation is also required based on the tank level or temperature data. The final compensated replenishment demand value will be more accurate than the actual water shortage, avoiding over- or under-replenishment caused by not considering thermal expansion and contraction. If the actual water shortage is 4.0L and the compensation correction is -0.3L, then the water replenishment requirement is 3.7L. After obtaining the water replenishment requirement, the system will execute a segmented automatic water replenishment control strategy to achieve a precise and safe water replenishment process. The core of segmented water replenishment is to compensate for the missing water through batch, multi-stage, and small-flow control, rather than replenishing a large amount at once. The system decomposes the water replenishment requirement into several replenishment stages, such as the start-up stage, steady-state replenishment stage, and convergence stage. In the initial stage of replenishment, the water replenishment valve is opened to a small opening to ensure that a rapid pressure rise does not trigger the safety valve or introduce new hydraulic shocks. When the replenished water reaches the estimated first-stage threshold, the system continues into the main replenishment stage. At this time, the water replenishment valve can be adjusted to a medium opening to maintain a stable flow rate. When the pressure gradually recovers and approaches the target pressure range (e.g., 0.23–0.25 MPa), the water replenishment enters the convergence stage, and the valve is lowered to a small opening again for a small amount of water replenishment, allowing the pressure to slowly approach the reference value.The system also monitors parameters such as pressure change rate, water replenishment duration, and temperature difference changes after water replenishment in real time to ensure a stable and reliable water replenishment process. After water replenishment is complete, the system automatically closes the water replenishment valve and re-enters normal monitoring mode, while recording the water replenishment behavior for subsequent big data analysis.

[0033] In this embodiment, the specific steps for performing segmented water replenishment control based on the water replenishment demand value to complete the water replenishment operation are as follows: Based on the water replenishment demand, segmented water replenishment calculations are performed to obtain the first segment water replenishment flow rate and the second segment water replenishment flow rate. Activate the water supply valve, perform water supply treatment based on the first water supply flow rate, and calculate the pressure change value of the expansion tank in real time during the water supply process; Calculate the rate of pressure rise of the pressure change value in the expansion tank; Based on the pressure rise rate, a pressure response effect analysis is performed. If the pressure rise rate is stable and linear, the water replenishment effect is considered normal. The water supply is gradually increased according to the second water supply flow rate to continuously replenish water. When the pressure value of the expansion tank is detected to be equal to the preset normal pressure reference value, the water replenishment operation is completed. When the pressure rate fluctuates abnormally, an abnormal warning signal is generated and uploaded to the cloud server, which is then marked as a system failure event.

[0034] In this embodiment, after obtaining the water replenishment demand, to ensure a smoother and safer replenishment process and avoid water hammer or severe system pressure fluctuations caused by excessive instantaneous replenishment, the overall water replenishment volume needs to be segmented. This step establishes a quantitative water replenishment model, decomposing the total water replenishment demand into a first-stage replenishment flow rate and a second-stage replenishment flow rate. Typically, the water replenishment demand is proportionally divided, for example, the first stage accounts for 20%–40% of the total replenishment volume, and the second stage accounts for 60%–80%. The specific proportion can be adjusted based on factors such as the heating system scale, pipeline capacity, and expansion tank structure. During the calculation process, the system combines parameters such as the flow characteristic curve of the water replenishment valve (a table showing the correspondence between valve opening and flow rate), water supply pressure, and water replenishment branch resistance to convert the water replenishment volume into the actual water replenishment flow rate. For example, if the water replenishment demand is 4 L, and the first stage plans to replenish 1 L and the second stage 3 L, then the required flow rate values ​​for the water replenishment valve at different stages are further calculated, such as 0.5 L / min for the first stage and 1.2 L / min for the second stage. This segmented strategy allows the water replenishment process to begin with a low flow rate, avoiding any pressure surges. The flow rate is then gradually increased after the system pressure stabilizes, improving replenishment efficiency. After determining the initial replenishment flow rate, the system begins the initial replenishment operation. The control system issues a command to open the replenishment valve to the corresponding set low flow rate, typically controlled between 5% and 15%, to ensure a smooth replenishment process. As the replenishment valve is activated, the system immediately enters real-time monitoring mode, continuously collecting pressure data from the expansion tank. The sampling period is generally 1 second or less to capture subtle pressure changes caused by the replenishment. As the replenishment volume gradually increases, the pressure inside the expansion tank rises accordingly. However, due to the low initial replenishment flow rate, the pressure change typically exhibits a slow and uniform growth trend. The system records the pressure data from each sample as a sequence, for example, continuously recording the trend from 0.21 MPa → 0.213 MPa → 0.216 MPa, and simultaneously adding a timestamp for subsequent analysis of the pressure change rate. During the process, the pressure signal will also undergo real-time noise removal and basic smoothing, such as filtering out certain brief pressure fluctuations, to ensure that the overall trend of pressure changes is accurately captured.

[0035] After obtaining continuously recorded pressure change data during the water replenishment process, quantitative calculations are needed to determine the actual impact of the water replenishment on system pressure. This step calculates the pressure rise rate, i.e., the increase in pressure per unit time. The calculation method typically uses differential or moving average methods, for example, dP / dt = [P(t) - P(t-1)] / Δt, where Δt is generally 1 second. If the pressure change sequence is 0.210→0.213→0.216→0.218 MPa, the pressure rise rates can be calculated as 0.003 MPa / s, 0.003 MPa / s, and 0.002 MPa / s, respectively. In the initial stage of water replenishment, due to the small flow rate, the pressure change rate is usually not too large, possibly falling within the range of 0.001–0.005 MPa / s. The system also performs short-term smoothing of the pressure rate, such as averaging the rate over the past 3–5 seconds, to avoid interference from single-point fluctuations. The rate of pressure change clearly indicates whether water replenishment is proceeding normally: a consistently positive rate indicates that water replenishment is effectively increasing system pressure; a rate of 0 or negative may indicate a blocked water replenishment path, insufficient water supply, an incompletely open water supply valve, or persistent leakage. After obtaining the pressure rise rate, the system needs to analyze its trend in depth to determine whether the water replenishment process conforms to normal hydraulic response characteristics. Under normal circumstances, the initial small-flow water replenishment will cause the expansion tank pressure to rise at a relatively stable rate, and its rate of change should be roughly linear, meaning that the pressure increase rate does not change significantly within a certain time range, and the curve slope is stable. When the pressure rise rate remains relatively stable over a continuous period, for example, consistently between 0.002 and 0.004 MPa / s, without significant fluctuations, the system determines that the water replenishment is normal, and the replenished water effectively increases system pressure. If the pressure rise rate fluctuates significantly, such as exhibiting unstable behavior like 0.003→0.000→0.004→0.001 MPa / s, it may indicate insufficient water replenishment, excessive water replenishment resistance, or ongoing pressure release within the system. Furthermore, if the pressure rise rate is non-linear, such as a sudden jump in rate (exceeding 0.006 MPa / s), it may indicate excessively rapid water intake, posing a risk of pressure surge. Once the first stage of water replenishment is deemed effective, the system enters the second stage of continuous water replenishment. The water replenishment volume in this stage is based on the setpoint for the second stage, typically using a higher flow rate than the first stage. For example, the valve opening may be increased from 10% in the first stage to 30%–50% to improve replenishment efficiency. During this process, the system continues to monitor the expansion tank pressure in real time and appropriately regulate the pressure rise rate to ensure the pressure remains within a safe range. As water replenishment continues, the expansion tank pressure will gradually approach the target normal pressure reference value.For example, if the reference value is set to 0.25 MPa, as the monitored pressure gradually rises to 0.24 MPa, 0.245 MPa, 0.248 MPa, etc., the system will gradually reduce the water supply or decrease the valve opening to allow the pressure to approach the target value more gently, preventing the pressure from exceeding the upper limit. When the expansion tank pressure stabilizes at the preset normal pressure reference value, the system determines that water replenishment is complete and immediately closes the water supply valve, entering the pressure maintenance monitoring phase. After water replenishment is completed, the system will continue to observe the pressure stability for a short period of time. If it is confirmed that the pressure does not drop rapidly, it will further rule out insufficient water replenishment or continuous leakage, thereby ensuring the safe and accurate completion of the water replenishment operation.

[0036] In this embodiment, the specific steps for generating an abnormal early warning signal and uploading it to the cloud server when the pressure rate fluctuates abnormally, and marking it as a system failure event, are as follows: When the pressure rate fluctuates abnormally, mark the abnormal information. The abnormal fluctuations include slow rises, pressure drops, and pressure rebounds followed by another drop. Generate abnormal early warning signals based on abnormal information; Record abnormal characteristic parameters, including the time point of abnormal pressure response, the deviation value of the relationship between water supply flow and pressure rise, and the rate of second pressure drop; Abnormal characteristic parameters and abnormal early warning signals are uploaded to the cloud server and marked as system failure events.

[0037] In this embodiment, during the water replenishment process, after the system continuously monitors the pressure changes in the expansion tank, if it detects a significant deviation of the pressure change rate from the normal linear upward trend, it needs to be identified as an abnormal pressure fluctuation and marked. Such abnormal fluctuations typically include three situations: First, a slow pressure increase, meaning that while maintaining a normal water replenishment flow rate, the pressure increase rate remains below the set minimum threshold for an extended period, for example, consistently below 0.0005 MPa / s, indicating that the water replenishment effect has failed to effectively increase the system pressure; second, a pressure drop, meaning the pressure value decreases instead of increasing during the water replenishment process, possibly from 0.225 MPa to 0.222 MPa, reflecting a leak path within the system or that the replenished water has not entered the main circulation pipeline; third, a brief rebound followed by a drop, for example, rising from 0.226 MPa to 0.229 MPa and then suddenly falling back to 0.223 MPa, this trend usually points to intermittent leakage or backpressure anomalies within the water replenishment system. When the system detects any of the above types of anomalies, it will immediately generate an anomaly marker and enter the anomaly warning logic. The early warning system will automatically generate abnormal early warning signals based on abnormal nodes, such as "abnormal water supply pressure response" or "possible continuous leakage" as prompts, and record key abnormal characteristic parameters, including the time point of the abnormality (e.g., 32 seconds after water supply), the deviation between the corresponding water supply flow rate and the pressure change trend at that time, and the rate of the second pressure drop, such as d. 2 P / dt 2 Negative values ​​exceeding a set threshold (e.g., -0.002 MPa / s) 2 All recorded abnormal characteristic parameters are packaged together and uploaded to the cloud server via wireless network or industrial communication module. After receiving this data, the cloud will mark such abnormal events as system failure events and create an abnormal record in the equipment asset file.

[0038] In this embodiment, an intelligent water replenishment control device for a heating system is provided, used to execute the intelligent water replenishment control method for a heating system as described above, including: The temperature difference calculation module is used to calculate the temperature difference between the supply and return water in the heating system and the standardized enthalpy difference sequence. The reference module is used to collect the rated configuration information of the heating system currently set, and to calculate the theoretical temperature difference reference value based on the rated configuration information; The deviation analysis module is used to perform temperature difference deviation analysis on the theoretical temperature difference benchmark value based on the supply and return water temperature difference value and the standardized enthalpy difference sequence, and to mark abnormal heat loss events. The pressure assessment module is used to extract the timestamps of abnormal events based on abnormal heat loss events; to perform pressure stability assessment for the corresponding time period based on the timestamps of abnormal events; and to determine that the heating system is in a leaking state when the pressure change rate gradient is detected to be gradually decreasing, and to enter the leak replenishment mode. The water replenishment control module is used to calculate the water replenishment demand value based on the water leakage replenishment mode; and to perform segmented water replenishment control according to the water replenishment demand value to complete the water replenishment operation.

[0039] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0040] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for intelligent water replenishment control of a heating system, characterized in that, Includes the following steps: Step S1: Calculate the temperature difference between the supply and return water in the heating system and the standardized enthalpy difference sequence; Step S2: Collect the rated configuration information of the heating system currently set, and calculate the theoretical temperature difference benchmark value based on the rated configuration information; Step S3: Perform temperature difference deviation analysis on the theoretical temperature difference benchmark value based on the supply and return water temperature difference value and the standardized enthalpy difference sequence, and mark abnormal heat loss events; Step S4: Extract the timestamp of the abnormal event based on the abnormal heat loss event; Based on the timestamp of the abnormal event, the pressure stability assessment is performed for the corresponding time period. When a gradual decrease in the pressure change rate gradient is detected, the heating system is determined to be in a leaking state and enters the leak replenishment mode. Step S5: Calculate the water replenishment demand value based on the water leakage replenishment mode; perform segmented water replenishment control according to the water replenishment demand value to complete the water replenishment operation.

2. The intelligent water replenishment control method for a heating system according to claim 1, characterized in that, The specific steps of step S1 are as follows: Real-time temperature parameters of the heating system's supply and return water pipes are collected using an online sensor array. Calculate the temperature difference of the real-time temperature parameters to obtain the supply and return water temperature difference; Real-time water flow data of the water supply and return pipelines are collected using electromagnetic flowmeters to obtain time-series flow curves. Instantaneous heat power transfer is calculated based on the time-series flow rate curve and the real-time temperature parameters to generate a supply and return water enthalpy difference sequence. The time-domain noise filter was applied to the enthalpy difference sequence of the supply and return water to obtain a standardized enthalpy difference sequence.

3. The intelligent water replenishment control method for a heating system according to claim 1, characterized in that, The specific steps of step S2 are as follows: Collect the current rated configuration information of the heating system, including the boiler rated power, design supply and return water temperature and rated circulation flow rate; Obtain outdoor ambient temperature and thermal parameters of the building envelope; Transient heat load is calculated based on the outdoor ambient temperature and the thermal parameters of the building envelope to obtain the instantaneous heat load demand; The theoretical supply and return water temperature difference is predicted based on the rated configuration information and instantaneous heat load demand to obtain the theoretical temperature difference benchmark value.

4. The intelligent water replenishment control method for a heating system according to claim 1, characterized in that, Step S3 is as follows: The supply and return water temperature difference is compared with the theoretical temperature difference benchmark value in real time to calculate the instantaneous temperature difference deviation. The instantaneous temperature difference deviation is fitted over time to obtain a deviation parameter set. Based on a preset supply and return water deviation threshold, a sliding window statistical analysis is performed on the deviation parameter set to mark the time points that exceed the supply and return water deviation threshold. Based on the time points, the standardized enthalpy difference sequence is synchronously matched to extract the enthalpy difference information at the corresponding time points; Calculate the rate of change of enthalpy difference and the second derivative based on enthalpy difference information; Trend fluctuation analysis is performed based on the rate of change of enthalpy difference and the second derivative. When the trend fluctuation of enthalpy difference is abnormal, it is determined to be an abnormal heat loss event.

5. The intelligent water replenishment control method for a heating system according to claim 1, characterized in that, The specific steps of step S4 are as follows: Real-time calculation of pressure monitoring data in the expansion tank; The pressure monitoring data is digitally filtered to obtain optimized pressure data; Extracting timestamps of abnormal events based on abnormal heat loss events; The pressure data is time-matched based on the timestamps of abnormal events, and the pressure change rate within that time period is calculated. Pressure stability is assessed based on the rate of pressure change. When a gradual decrease in the rate of pressure change is detected, the heating system is determined to be in a leaking state and enters the leak replenishment mode.

6. The intelligent water replenishment control method for a heating system according to claim 1, characterized in that, The specific steps of step S5 are as follows: Based on the water leakage replenishment mode, the difference between the pressure monitoring data and the preset normal pressure reference value is calculated to obtain the water tank pressure loss value; The total heat loss of the system is calculated based on the supply and return water temperature difference and the standardized enthalpy difference sequence. The actual water shortage is calculated by back-calculating the missing water tank pressure value and the total heat loss of the system. The actual water shortage is corrected by pipeline expansion compensation to obtain the water replenishment demand value; Segmented water replenishment control is performed based on the water replenishment demand value to complete the water replenishment operation.

7. The intelligent water replenishment control method for a heating system according to claim 6, characterized in that, The specific steps for performing segmented water replenishment control based on the water replenishment demand value to complete the water replenishment operation are as follows: Based on the water replenishment demand, segmented water replenishment calculations are performed to obtain the first segment water replenishment flow rate and the second segment water replenishment flow rate. Activate the water supply valve, perform water supply treatment based on the first water supply flow rate, and calculate the pressure change value of the expansion tank in real time during the water supply process; Calculate the rate of pressure rise of the pressure change value in the expansion tank; Based on the pressure rise rate, a pressure response effect analysis is performed. If the pressure rise rate is stable and linear, the water replenishment effect is considered normal. The water supply is gradually increased according to the second water supply flow rate to continuously replenish water. When the pressure value of the expansion tank is detected to be equal to the preset normal pressure reference value, the water replenishment operation is completed. When the pressure rate fluctuates abnormally, an abnormal warning signal is generated and uploaded to the cloud server, which is then marked as a system failure event.

8. The intelligent water replenishment control method for a heating system according to claim 7, characterized in that, The specific steps for generating an abnormal early warning signal and uploading it to the cloud server when the pressure rate fluctuates abnormally, and marking it as a system failure event, are as follows: When the pressure rate fluctuates abnormally, mark the abnormal information. The abnormal fluctuations include slow rises, pressure drops, and pressure rebounds followed by another drop. Generate abnormal early warning signals based on abnormal information; Record abnormal characteristic parameters, including the time point of abnormal pressure response, the deviation value of the relationship between water supply flow and pressure rise, and the rate of second pressure drop; Abnormal characteristic parameters and abnormal early warning signals are uploaded to the cloud server and marked as system failure events.

9. An intelligent water supply control device for a heating system, characterized in that, The method for implementing the intelligent water replenishment control of the heating system as described in claim 1 includes: The temperature difference calculation module is used to calculate the temperature difference between the supply and return water in the heating system and the standardized enthalpy difference sequence. The reference module is used to collect the rated configuration information of the heating system currently set, and to calculate the theoretical temperature difference reference value based on the rated configuration information; The deviation analysis module is used to perform temperature difference deviation analysis on the theoretical temperature difference benchmark value based on the supply and return water temperature difference value and the standardized enthalpy difference sequence, and to mark abnormal heat loss events. The pressure assessment module is used to extract the timestamps of abnormal events based on abnormal heat loss events; to perform pressure stability assessment for the corresponding time period based on the timestamps of abnormal events; and to determine that the heating system is in a leaking state when the pressure change rate gradient is detected to be gradually decreasing, and to enter the leak replenishment mode. The water replenishment control module is used to calculate the water replenishment demand value based on the water leakage replenishment mode; and to perform segmented water replenishment control according to the water replenishment demand value to complete the water replenishment operation.

Images