Remote monitoring campus hot water equipment fault early warning method and system

Abstract

The invention provides a remote monitoring campus hot water equipment fault early warning method and system, and the method comprises the steps: deploying a sensor, collecting the real-time data of campus hot water equipment in real time, and recognizing the current working state of the equipment; dynamically carrying out fault detection according to the current working state of the equipment, and generating a fault detection signal; according to the fault detection signal, monitoring and determining whether the equipment has the problem of efficiency loss or energy abnormity by using thermodynamic-based energy conservation analysis, and outputting a comprehensive fault signal; designing a fault scoring model and a grading early warning mechanism according to the comprehensive fault signal to carry out grading early warning on the operation state of the equipment; and designing an adaptive adjustment mechanism to perform long-term optimization on the fault scoring model. The accuracy and the real-time performance of fault early warning of the campus hot water equipment are greatly improved, and a powerful technical guarantee is provided for efficient operation of a campus heat supply system.

Description

Technical Field

[0001] The present invention belongs to the technical field of fault warning, and in particular relates to a remotely monitored campus hot water equipment fault warning method and system. Background Art

[0002] With the continuous development of campus infrastructure, the wide application of hot water equipment in dormitory buildings, teaching buildings, canteens and other places has provided convenience for the daily life of teachers and students. Traditional hot water equipment management relies more on manual maintenance and regular inspections. The operating status and potential faults of the equipment are often difficult to detect in time, which can easily cause sudden failures, leading to campus heating interruption and affecting normal study and life. At the same time, the operating data of traditional hot water equipment (such as temperature, water pressure, heating power, etc.) are relatively scattered, lacking unified management and real-time monitoring methods, and it is difficult to form a complete fault warning system. Most of the existing remote monitoring technologies rely on a simple threshold alarm mechanism, that is, once a sensor data exceeds the preset range, an alarm is triggered. However, this method has obvious limitations: First, the single threshold alarm method cannot take into account the different working states of the equipment (such as heating, insulation, idle, etc.), so that at different stages of equipment operation, the accuracy and sensitivity of the warning are biased, and false alarms or missed alarms often occur. Secondly, since the operation of hot water equipment involves complex physical processes, it is difficult for existing technologies to perform energy conservation analysis based on thermodynamic principles, and thus it is impossible to accurately judge the heating efficiency, energy consumption and equipment aging of the equipment. Potential problems. Most existing monitoring systems are based on statistical analysis and lack a deep understanding of the working principles of the equipment, and are unable to identify hidden faults caused by problems such as aging of heating elements or water pressure fluctuations.

[0003] In addition, most campus hot water equipment is installed in a distributed manner, and traditional monitoring systems often rely on centralized servers for data aggregation and analysis. This method not only increases the delay in data transmission, but also easily leads to single point failures. Once some hot water equipment or monitoring systems have problems, the fault information may not be transmitted in time, missing the best maintenance time. With the increase in the number of devices, the processing capacity and analysis efficiency of traditional monitoring methods are also limited when dealing with large amounts of sensor data, making it difficult to ensure the scalability and real-time performance of the system. Another drawback of existing remote monitoring technology is that it is usually limited to alarms after equipment failures occur, and lacks effective prediction and early warning capabilities. It is impossible to identify potential equipment failures in advance, resulting in equipment maintenance often falling into a "passive maintenance" situation, which not only increases maintenance costs, but may also cause equipment damage and downtime due to failures not being handled in a timely manner.

[0004] In summary, although the existing technology can realize remote monitoring and partial early warning functions of hot water equipment, it still has significant deficiencies in handling complex operating conditions, improving the accuracy of fault detection, and predicting potential faults in advance. These deficiencies make it difficult for existing systems to cope with the long-term high-load operation requirements of hot water equipment in complex campus environments. Therefore, it is particularly important to design a system that can adaptively adjust the early warning mechanism according to the equipment status and perform intelligent analysis based on the physical properties of the equipment (such as thermodynamic energy conservation). Such a system can not only improve the accuracy and real-time performance of fault detection, but also provide effective technical support for predictive maintenance of equipment, avoiding unnecessary losses due to failure to detect faults in time. Summary of the invention

[0005] The purpose of the present invention is to design a remotely monitored campus hot water equipment fault warning method and system to effectively solve the problems existing in the prior art, such as difficulty in identifying equipment status, insufficient fault warning sensitivity, and inability to perform energy efficiency analysis.

[0006] In order to achieve the above object, a first aspect of the present invention provides a remotely monitored campus hot water equipment failure early warning method, comprising the following steps: Deploy sensors to collect real-time data from campus hot water equipment and identify the current working status of the equipment; Dynamically perform fault detection based on the current working status of the equipment and generate a fault detection signal; According to the fault detection signal, use the energy conservation analysis based on thermodynamics to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems, and output a comprehensive fault signal; Design a fault scoring model and a graded warning mechanism based on comprehensive fault signals to provide graded warnings for equipment operation status; Design an adaptive adjustment mechanism to optimize the fault scoring model in the long term; After the real-time data of the campus hot water equipment is collected in real time, a state transition model is constructed to predict the current state according to the current sensor reading and the state history, and the predicted current state is used as the identified current state, specifically: Assume that the state at the previous moment is , define the state transition probability , and dynamically adjusted according to the sensor data, the state transition of the state transition model is expressed as follows: ; in, Indicates current data, Indicates water temperature data, represents water pressure data, t represents time t, Represents a fixed time interval, Indicates the current state. For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state Make estimates; The state transition rules are defined as follows: ; ; ; Among them, if the current Greater than zero and the water temperature rise rate Greater than the set threshold When the device is in heating state If the current Near zero and water temperature Keep stable, that is , the equipment is in the insulation state If the current If the temperature is zero and the water temperature gradually decreases, the device is in idle state. ; At the same time, a memory decay factor is introduced in the state transfer model , weighted calculation of state history is performed as follows: ; in, represents the transition probability based on historical state data, represents the transition probability calculated by the current sensor data, Control the balance of historical and current data.

[0007] Furthermore, the current working state of the device depends on the data output by the sensor: water temperature data, water pressure data and current data; The dynamically performing fault detection according to the current working state of the device and generating a fault detection signal specifically includes: In the heating state Perform fault detection under In the insulation state Perform fault detection under In the idle state Perform fault detection under The heating state , insulation state and idle state Comprehensive fault detection signal .

[0008] Further, in the heating state Fault prediction is performed as follows: ; in, Indicates the heating efficiency, represents the rate of change of temperature, Indicates current data, is the energy loss coefficient, which is related to the equipment water pressure The fluctuation of If the heating efficiency Below the set minimum heating efficiency threshold , then the heating element has an inefficient fault; if the water pressure fluctuates greatly, that is Greater than a preset value , then increase value, conduct heating efficiency analysis; when abnormal heating efficiency is detected, a fault signal is generated , with heating element status information; In the insulation state The fault detection is performed as follows: ; in, is the water temperature change, if ,in is the threshold value of water temperature fluctuation allowed, If it is the insulation energy conservation term, it is judged that the equipment has poor insulation or abnormal heat dissipation, and a fault signal is generated ; In the idle state The fault detection is performed as follows: ; Among them, in the idle state, it is determined whether there is abnormal current leakage, that is, Or the water temperature does not drop normally, i.e. If there is current leakage or the water temperature drops slowly, it is judged as a fault and a signal is generated. ; The heating state , insulation state and idle state Comprehensive fault detection signal , which is expressed as follows: ; in, , , Represent the fault detection signals in heating, insulation and idle states respectively.

[0009] Furthermore, according to the fault detection signal, the energy conservation analysis based on thermodynamics is used to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems, and output a comprehensive fault signal, which is divided into heating state, insulation state and idle state.

[0010] Further, the energy calculation in the heating state is expressed as follows: Heating status First, calculate the power input of the device. Based on the current and voltage readings of the device, the power input Calculated as: ; in, is the electrical energy input during heating, is the voltage, is the current value, is the time interval; Calculate the heat output during the heating process by changing the water temperature : ; in, is the quality of the water, is the specific heat capacity of water, Indicates the rate of change of water temperature ; Energy difference Reflects the degree of energy loss: ; if If the value is too large, it indicates that there is energy loss in the heating process, which is due to aging of the heating elements or reduced system energy efficiency; An innovative regularization term is introduced , which is used to compensate for the energy change caused by external pressure fluctuations, is expressed as follows: ; in, represents energy loss, is the adjustment factor, is the measured value of the current water pressure; if Exceeding a predetermined threshold , the system generates a heating energy loss fault signal , indicating that there is a serious energy loss during the heating process; In the insulation state Under temperature change Exceeding the set threshold , and no current input , there is energy leakage or abnormal heat dissipation: ; in, It is the temperature fluctuation threshold during the insulation stage. When the temperature fluctuation of the device exceeds this value and there is no additional energy input, it is judged that there is a fault in the insulation state of the device, and a insulation state loss fault signal is generated. ; In idle state The device should stop working and the current should be zero; if the current Greater than a certain threshold , it indicates that there is an electrical fault or current leakage in the device: ; At the same time, the water temperature drop rate is monitored in real time. If the water temperature does not drop significantly, , there is a poor heat dissipation problem, and a corresponding fault signal is generated .

[0011] Furthermore, the comprehensive fault signal is expressed as follows: ; in, Indicates a comprehensive fault signal.

[0012] Furthermore, the fault scoring model combines energy loss And the fault frequency are used to calculate the fault score, which is expressed as follows: ; in, Rate the fault, and are weighted coefficients, respectively controlling energy loss With fault signal , the impact on the ratings; It is the adjustment coefficient under different fault states to ensure that the influence weight of each state on the fault score is reasonable.

[0013] Furthermore, an early warning regularization term is introduced into the fault scoring model. , the effect of device status change in the corrected scoring model is expressed as follows: ; in, represents the corrected fault score, Used to adjust the warning sensitivity of the equipment under high load or low load conditions; The hierarchical early warning mechanism includes: Low-level warning: , indicating that the equipment is operating basically normally, with slight energy loss or occasional failures, and regular maintenance is recommended; Intermediate warning: , indicating that energy loss or fault signals frequently occur during equipment operation, which poses a potential risk. It is recommended to troubleshoot the problem in advance; Advanced warning: , indicating that a serious fault or long-term energy abnormality has occurred during the operation of the equipment, and the equipment must be shut down for inspection immediately to avoid further deterioration of the system.

[0014] Furthermore, the design of the adaptive adjustment mechanism performs long-term optimization on the fault scoring model, specifically including: Adaptively adjust the energy loss weight according to the current corrected fault score of the device; Adjust the fault frequency weight according to the fault signals under different equipment states; Introduce feedback regularization term to compensate for the fault detection deviation caused by external disturbances; By analyzing the changing trend of fault scores, the fault scoring model is optimized in the long term; The energy loss weight, fault frequency weight and feedback regularization term are optimized in a closed loop by analyzing their changing trends.

[0015] In a second aspect of the present invention, a remotely monitored campus hot water equipment failure early warning system is provided, the system comprising: The equipment data collection module is used to deploy sensors to collect real-time data of campus hot water equipment and identify the current working status of the equipment; A fault signal initialization module is used to dynamically perform fault detection according to the current working status of the equipment and generate a fault detection signal; A fault signal generation module is used to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems based on the fault detection signal using energy conservation analysis based on thermodynamics, and output a comprehensive fault signal; The fault warning module is used to design a fault scoring model and a graded warning mechanism based on comprehensive fault signals to provide graded warnings for the equipment operation status; System optimization module, which is used to design an adaptive adjustment mechanism to optimize the fault scoring model in the long term; After the real-time data of the campus hot water equipment is collected in real time, a state transition model is constructed to predict the current state according to the current sensor reading and the state history, and the predicted current state is used as the identified current state, specifically: Assume that the state at the previous moment is , define the state transition probability , and dynamically adjusted according to the sensor data, the state transition of the state transition model is expressed as follows: ; in, Indicates current data, Indicates water temperature data, represents water pressure data, t represents time t, Represents a fixed time interval, Indicates the current state. For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state Make estimates; The state transition rules are defined as follows: ; ; ; Among them, if the current Greater than zero and the water temperature rise rate Greater than the set threshold When the device is in heating state If the current Near zero and water temperature Keep stable, that is , the equipment is in the insulation state If the current If the temperature is zero and the water temperature gradually decreases, the device is in idle state. ; At the same time, a memory decay factor is introduced in the state transfer model , weighted calculation of state history is performed as follows: ; in, represents the transition probability based on historical state data, represents the transition probability calculated by the current sensor data, Control the balance of historical and current data.

[0016] The beneficial technical effects of the present invention are at least as follows: First, by introducing a state transition model, the present invention can dynamically adjust the fault detection strategy according to the different working states of the equipment (such as heating, heat preservation, idle, etc.). Compared with the traditional fixed threshold warning method, the state transition model can identify the different stages of equipment operation and set corresponding warning thresholds based on the characteristics of different stages. For example, in the heating state, the system pays more attention to the matching of heating power and water temperature rise rate; while in the heat preservation or idle state, the system monitors whether the energy consumption of the equipment is abnormal. Through this adaptive warning mechanism, the system can maintain efficient fault detection in different working states to avoid false alarms or missed alarms.

[0017] Secondly, the present invention introduces a thermodynamic anomaly detection model. Using the energy conservation principle of thermodynamics, the system can monitor the energy input and output of the equipment in real time and determine whether its heating efficiency and energy consumption are reasonable. Unlike the existing fault detection methods based on statistical models, the thermodynamic model can accurately identify potential faults caused by energy loss or reduced heating efficiency by analyzing the physical properties of the equipment. For example, the system can detect potential problems such as aging of heating elements or abnormal heat dissipation of equipment by comparing the power consumption and water temperature changes of the equipment. This not only detects hidden faults of the equipment earlier, but also avoids greater damage to the equipment due to long-term low energy efficiency.

[0018] In addition, the present invention also realizes real-time monitoring and dynamic early warning of equipment through remote monitoring and distributed early warning systems. The system can not only issue an early warning in the first time when a fault occurs, but also has a predictive maintenance function. It can predict the potential failure trend of equipment through the accumulation of long-term data, thereby providing effective equipment maintenance suggestions for campus managers and reducing the inconvenience caused by sudden shutdown and sudden failure of equipment. Through these innovations, the present invention greatly improves the accuracy and real-time nature of campus hot water equipment fault warnings, and provides a strong technical guarantee for the efficient operation of the campus heating system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0020] Figure 1 This is a flow chart of a remotely monitored campus hot water equipment fault warning method of the present invention. DETAILED DESCRIPTION

[0021] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0022] In one or more embodiments, Figure 1 As shown, a remote monitoring campus hot water equipment failure warning method is disclosed, and the method includes the following steps S1-S5: S1. Deploy sensors to collect real-time data of campus hot water equipment and identify the current working status of the equipment.

[0023] Specifically, the present invention installs multiple sensors on the hot water equipment to obtain real-time operation data. Key data include: Water temperature data :Real-time measurement of the outlet water temperature of the hot water equipment in degrees Celsius (℃).

[0024] Water pressure data : Real-time measurement of water pressure in Pascals ( ).

[0025] Current data : Measures the current during heating in amperes ( ).

[0026] Furthermore, the data collection frequency is set to a fixed time interval (For example, sampling once every minute) to form time series data. In order to deal with sensor noise, the sensor data is initially smoothed using a sliding average algorithm: ; in, is the water temperature after smoothing. is the size of the sliding window (e.g. ). Other sensor data and The same smoothing process is performed. is the counting variable in the sum, representing the time offset of each data point in the sliding window relative to t.

[0027] Furthermore, the sensor data characteristics of the device will be significantly different under different operating conditions. The specific status definitions are as follows: Heating status : Current Significantly increased water temperature rise.

[0028] Insulation status : Current Near zero water temperature Maintain stability.

[0029] Idle state : Current The water temperature gradually drops to zero.

[0030] To achieve dynamic state recognition, a state transition model is defined, which predicts the current state based on the current sensor readings and state history. . Assume that the state at the previous moment is , the present invention defines the state transition probability , and dynamically adjust according to sensor data. The state transition of the model is based on the following formula: ; in, For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state The state transition model assumes that the transition probabilities of the heating, insulation, and idle states depend on the changes in sensor data (e.g., the rate of water temperature rise). and the change of current , dynamically adjusted based on past data.

[0031] Furthermore, to determine the current state of the device, the threshold and change rate of the sensor data are combined: When the current Significantly greater than zero and the water temperature rise rate Greater than the set threshold When the system determines that the device is in the heating state .

[0032] When the current Near zero and water temperature Keep stable ( ), the equipment is in the insulation state .

[0033] When the current The system determines that the device is in idle state. .

[0034] The state transition rules defined are: ; ; ; in, It is a minimum value and is used to determine whether the water temperature is stable.

[0035] Furthermore, the transition probability in the state transition model The memory decay factor is dynamically adjusted according to the operation history of the device. , used to perform weighted calculation on state history: ; in, is the transition probability based on historical state data, is the transition probability calculated from the current sensor data. Control the balance between historical and current data, typical value .

[0036] Through this dynamic adjustment, the system can adapt to long-term operational changes of the equipment, such as the impact of seasonal changes or equipment aging on the state transition rules.

[0037] This step realizes accurate identification of the operating status of campus hot water equipment through real-time data collection, construction of state transition model and dynamically adjusted state judgment mechanism. The whole process is driven by sensor data and combined with the state transition law of historical data, it can dynamically adapt to different operating environments of the equipment.

[0038] The final output is the current status of the device (heating, keeping warm, idle) and corresponding sensor data , , ,These outputs will become the inputs for the subsequent fault detection steps.

[0039] S2. Dynamically perform fault detection according to the current working status of the equipment and generate a fault detection signal.

[0040] Specifically, in the heating state, the goal of the device is to quickly increase the water temperature. Therefore, fault detection mainly focuses on the heating efficiency, which is determined by the water temperature change. and current Definition of relationship.

[0041] Based on the balance relationship between physical energy input and output, the present invention defines a heating efficiency formula and introduces an innovative energy loss regularization term , used to detect unstable factors in the energy transmission process, such as energy loss caused by equipment aging.

[0042] ; in, Indicates the heating efficiency, is the rate of change of temperature, is the current, is the energy loss coefficient, which is related to the equipment water pressure The fluctuation of is related. The purpose is to compensate for equipment efficiency losses due to water pressure fluctuations or aging of heating elements.

[0043] Furthermore, the fault detection logic is expressed as follows: calculate ,if Below the set minimum heating efficiency threshold , the system determines that the heating element has an inefficient fault; If the water pressure fluctuates greatly ( Greater than a preset value ), the system increases value, and conduct a more detailed analysis of heating efficiency; When the system detects abnormal heating efficiency, a fault signal is generated , with heating element status information.

[0044] Furthermore, in the insulation state, the goal of the device is to keep the water temperature stable and minimize energy consumption. Therefore, the present invention focuses on monitoring the fluctuation of temperature and current. , which is used to prevent unstable energy transfer or difficulty in maintaining temperature when the equipment is in a heat preservation state. The detection strategy is analyzed by the following formula: ; like (in is the threshold value of water temperature fluctuation), the system determines that the device may have poor insulation or abnormal heat dissipation. At this time, a fault signal is generated .

[0045] Understandably, in the heat preservation state, It is related to the energy consumption of the equipment and the external environmental conditions. If the internal energy regulation of the equipment is unstable or the ambient temperature fluctuates too much, The detection sensitivity is dynamically adjusted to suit the actual operating conditions.

[0046] Furthermore, in the idle state, the device should no longer consume power and the water temperature should drop naturally. The core monitoring at this time is current leakage detection and temperature drop rate detection. The detection formula is as follows: ; Among them, in the idle state, the system determines whether there is abnormal current leakage ( ) or the water temperature does not drop normally ( ). If there is a current leakage or the water temperature drops slowly, the system determines it as a fault and generates a signal .

[0047] Final fault detection signal is the sum of fault signals in all states: ; in, , , Represents the fault detection signals in heating, insulation and idle states respectively, It will be provided to the energy anomaly analysis module as input for subsequent steps for further verification and early warning.

[0048] Furthermore, during the entire fault detection process, the system will feedback and adjust the future state transition model based on historical fault information and the detection results of the current state. If false alarms or missed alarms occur frequently in a certain state, the system will automatically adjust the fault detection parameters of the relevant state (such as , and ) to improve the accuracy of future detection.

[0049] This step realizes multi-state real-time fault detection of hot water equipment by introducing an innovative state-related fault detection mechanism, and compensates for the shortcomings of the traditional model in energy loss through regularization terms. It will be used for the next step of energy anomaly analysis, providing a strong foundation for the system's high-precision early warning.

[0050] S3. Based on the fault detection signal, use the energy conservation analysis based on thermodynamics to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems, and output a comprehensive fault signal.

[0051] Specifically, in the heating state First, calculate the power input of the device. Based on the current and voltage readings of the device, the power input Calculated as: ; in, is the electrical energy input during heating, is the voltage, is the current value, is the time interval.

[0052] Next, calculate the heat output during the heating process by changing the water temperature : ; in, is the quality of the water, is the specific heat capacity of water, Indicates the rate of change of water temperature .

[0053] Energy difference Reflects the degree of energy loss: ; if A value that is too high indicates that there is energy loss in the heating process, possibly due to aging of the heating elements or reduced system efficiency.

[0054] It is understandable that since the equipment may be affected by water pressure fluctuations, heating instability or external environmental factors during operation, the analysis of energy loss needs to take external conditions into consideration. To this end, the present invention introduces an innovative regularization term , used to compensate for the energy changes caused by external pressure fluctuations: ; in, is the adjustment factor, is the measured value of the current water pressure. This regularization term is dynamically adjusted according to the long-term operation of the equipment. , to reflect the energy loss caused by equipment aging or environmental changes. Exceeding a predetermined threshold , the system generates a heating energy loss fault signal , indicating that there is serious energy loss during the heating process.

[0055] Furthermore, in the heat preservation state In this case, the system focuses on monitoring water temperature changes and energy input. Exceeding the set threshold , and no current input , there may be energy leakage or abnormal heat dissipation: ; in, It is the temperature fluctuation threshold during the insulation phase. When the device's temperature fluctuation exceeds this value and there is no additional energy input, the system determines that the device insulation state is faulty.

[0056] Furthermore, in idle state The device should stop working and the current should be zero. Greater than a certain threshold , it indicates that the device may have an electrical fault or current leakage: ; At the same time, the system monitors the rate of temperature drop. If the water temperature does not drop significantly ( ), there may be a poor heat dissipation problem, generating a corresponding fault signal .

[0057] Comprehensive energy conservation analysis and fault signal output Combining the above steps, the comprehensive fault signal of energy conservation analysis is It consists of the fault signal and energy loss regularization term in each state: ; This signal is used in the subsequent fault warning stage, indicating that the equipment has energy anomalies during the heating, insulation or idle stages.

[0058] It is understandable that during the entire energy conservation analysis process, the system will adjust the regularization term according to the energy anomaly detection results of the device. Adaptive adjustments are made to gradually optimize model parameters to accommodate changes in the state of the equipment during long-term operation. The feedback adjustment process ensures that the model can cope with equipment aging and seasonal changes.

[0059] This step achieves accurate energy anomaly detection and fault signal generation by calculating the thermodynamic energy input and output, introducing innovative regularization terms, and combining the energy conservation characteristics of the device in different states. This process provides a reliable basis for equipment fault detection and early warning, and helps improve the stability of system operation and the accuracy of fault identification.

[0060] S4. Design a fault scoring model and a graded warning mechanism based on the comprehensive fault signal to provide graded warning for the equipment operation status.

[0061] Specifically, in order to better quantify the severity of equipment failures, the system designed an innovative fault scoring model that calculates the fault score by combining energy loss and fault frequency. , the formula is as follows: ; in, Rate the fault, and are weighted coefficients, respectively controlling energy loss With fault signal (Heating failure , insulation failure , Idle Fault ) on the ratings. The adjustment coefficients for different fault states ensure that the weight of each state on the fault score is reasonable. The fault score measures the health of the equipment by combining the energy loss during equipment operation (for example, too much electrical energy cannot be converted into heat energy) with the frequency of equipment failures. By weighting the fault conditions under each state, the system can dynamically determine the risk level of equipment failure.

[0062] Furthermore, for fault detection of equipment in complex operating environments, the system introduces an innovative early warning regularization term , which is used to correct the impact of device state changes in the scoring model. This regularization term is dynamically adjusted based on device load, operating time and historical data. The specific calculation formula is: ; in, Used to adjust the warning sensitivity of the equipment under high load or low load conditions. For example, when the equipment is in a high load operation state for a long time, the system may relax the sensitivity to small faults, while when the equipment is in a low load state, the regularization term will increase the system's detection sensitivity to energy anomalies and small faults to prevent potential problems from being ignored.

[0063] Furthermore, the system adjusts the fault score based on Provide graded warnings for equipment failures. According to the size of the failure score, the warnings are divided into the following three levels: Low-level warning: The equipment is operating normally, but there may be slight energy loss or occasional failures. Regular maintenance is recommended.

[0064] Intermediate warning: Energy loss or fault signals frequently occur during equipment operation, posing potential risks. It is recommended to troubleshoot the problem in advance.

[0065] Advanced warning: . It means that a serious fault or long-term energy abnormality has occurred during the operation of the equipment, and the equipment must be shut down for inspection immediately to avoid further deterioration of the system.

[0066] Furthermore, the system is not only based on the fault score at the current moment Early warning is carried out, and the early warning result is further corrected by analyzing the time change trend of the fault signal. , which is used to calculate the difference between the current fault score and the score at the previous moment: ; if The fault score gradually increases, indicating that the fault is getting worse. The system will further increase the warning level based on the original warning. When it decreases, it means that the risk of failure is easing, and the system will appropriately relax the warning sensitivity to avoid excessive alarms.

[0067] It is understandable that the system continuously optimizes the parameters in the fault scoring model through an adaptive feedback mechanism. If the energy loss has a significant impact on the system operation, the system will increase value, so that The weight in the score increases; if the frequency of fault signals is more significant, the system will increase it appropriately. and , enhancing the impact of fault signals on the score.

[0068] This feedback mechanism ensures that the model is highly adaptable to the operating status of the equipment and can respond promptly to complex situations such as environmental changes, load changes, and equipment aging.

[0069] Further, the final warning signal The output format is: ; explain: The final warning signal, including the current fault score , warning level (low, medium or high), fault type (heating, insulation, idle), current device status Based on the warning level, the system will automatically generate corresponding maintenance suggestions, such as regular inspection (low level), early problem detection (medium level) or immediate shutdown for maintenance (high level).

[0070] Furthermore, the early warning signal is transmitted to the management center through the remote monitoring platform to ensure that the maintenance team can obtain fault information in a timely manner. According to different early warning levels, maintenance suggestions include different levels of operation instructions such as regular inspection, emergency repair and equipment replacement, helping managers make the best maintenance decisions.

[0071] This step uses an innovative fault scoring model and warning regularization term, combined with energy loss and fault frequency, to dynamically implement warning classification during equipment operation. The graded warning mechanism not only improves the system's sensitivity to changes in equipment status, but also adaptively adjusts system parameters through a feedback optimization mechanism, providing a clear decision-making basis for equipment maintenance. The output of the warning signal ensures that managers can make corresponding maintenance measures based on actual conditions, effectively extending the service life of the equipment and reducing the risk of failures.

[0072] S5. Design an adaptive adjustment mechanism to optimize the fault scoring model in the long term.

[0073] Specifically, based on the current fault score of the device , the system weights energy loss Adaptive adjustments are made to ensure that fault scores better reflect energy anomalies: ; in, represents the weight of energy loss in the score, is the adjustment rate constant, Rate the current fault, Score the target. When it is higher, it indicates that the energy loss is significant and the system is improved. The weight of , ensures that energy loss accounts for a larger proportion in the score.

[0074] Furthermore, according to the fault signals under different states of the equipment, the system weights the fault frequency Adjustments are made to increase the system's sensitivity to frequent failures: ; represents the weight of the failure frequency, To adjust the rate constant, It is the fault signal increment in heating, insulation and idle state. ,When faults occur frequently, the system will pay more attention to the relevant states, and increase the impact of the fault frequency in the score.

[0075] Furthermore, the system introduces an innovative feedback regularization term , which is used to compensate for the fault detection deviation caused by external disturbances (such as water pressure fluctuations and temperature changes) of the equipment, and ensure the adaptability of the model to complex operating environments: ; is the regularization coefficient, is the current energy conservation fault signal, is the historical average energy signal. By adjusting , the system can adaptively respond to changes in external conditions and improve its ability to compensate for environmental disturbances.

[0076] Furthermore, the system analyzes the changing trend of fault scores. , optimize the model for the long term: ; if If it continues to rise, it means that the fault is getting worse, and the system will further increase the model's sensitivity to energy anomalies and fault frequency; If it decreases, it means that the fault is being alleviated and the system will reduce the warning sensitivity to avoid excessive alarms.

[0077] Furthermore, the system continuously accumulates the equipment's historical fault data, energy conservation signals, and feedback adjustment results to form equipment operation archives. Through retrospective analysis of historical data, the system can identify the equipment's operating mode and potential failure trends, thereby optimizing future maintenance plans. Through long-term data analysis, the system can generate a predictive maintenance model to identify potential equipment hazards in advance and further optimize the equipment's fault detection and maintenance efficiency.

[0078] Furthermore, the feedback mechanism of the system forms a closed loop through the following formula, gradually optimizing all key parameters (including , and ): ; represents the comprehensive set of model parameters, Through continuous feedback, the system's detection model is continuously optimized to ensure robustness and accuracy under the influence of equipment operating environment, load fluctuations or aging.

[0079] Furthermore, the system generates maintenance plans based on a closed-loop feedback mechanism and automatically gives intelligent maintenance suggestions for equipment based on predictive analysis of historical data, helping managers to develop reasonable maintenance and inspection plans, extend equipment life and reduce the risk of failure.

[0080] Through feedback and adaptive adjustment mechanisms, the system can not only adjust key detection parameters in real time, but also perform global optimization through long-term trend analysis to ensure that the equipment maintains efficient fault detection and maintenance capabilities under complex operating conditions. The continuous optimization of closed-loop feedback and adaptive models enhances the system's intelligent prediction and adaptability.

[0081] In one or more embodiments, a second aspect of the present invention discloses a remotely monitored campus hot water equipment failure warning system, the system comprising: The equipment data collection module 1011 is used to deploy sensors to collect real-time data of campus hot water equipment and identify the current working status of the equipment; A fault signal initialization module 1012 is used to dynamically perform fault detection according to the current working state of the device and generate a fault detection signal; The fault signal generating module 1013 is used to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems by using energy conservation analysis based on thermodynamics according to the fault detection signal, and output a comprehensive fault signal; Fault warning module 1014, used to design a fault scoring model and a graded warning mechanism according to the comprehensive fault signal to provide graded warning for the equipment operation status; The system optimization module 1015 is used to design an adaptive adjustment mechanism to optimize the fault scoring model in the long term; After the real-time data of the campus hot water equipment is collected in real time, a state transition model is constructed to predict the current state according to the current sensor reading and the state history, and the predicted current state is used as the identified current state, specifically: Assume that the state at the previous moment is , define the state transition probability , and dynamically adjusted according to the sensor data, the state transition of the state transition model is expressed as follows: ; in, Indicates current data, Indicates water temperature data, represents water pressure data, t represents time t, Represents a fixed time interval, Indicates the current state. For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state Make estimates; The state transition rules are defined as follows: ; ; ; Among them, if the current Greater than zero and the water temperature rise rate Greater than the set threshold When the device is in heating state If the current Near zero and water temperature Keep stable, that is , the equipment is in the insulation state If the current If the temperature is zero and the water temperature gradually decreases, the device is in idle state. ; At the same time, a memory decay factor is introduced in the state transfer model , weighted calculation of state history is performed as follows: ; in, represents the transition probability based on historical state data, represents the transition probability calculated by the current sensor data, Control the balance of historical and current data.

[0082] The above disclosures are only some preferred embodiments of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiments and equivalent changes made according to the claims of the present invention are still within the scope of the invention.

Claims

1. A remote monitoring campus hot water equipment failure early warning method, characterized in that: The following steps are involved: Deploy sensors to collect real-time data from campus hot water equipment and identify the current working status of the equipment; Dynamically perform fault detection based on the current working status of the equipment and generate a fault detection signal; According to the fault detection signal, use the energy conservation analysis based on thermodynamics to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems, and output a comprehensive fault signal; Design a fault scoring model and a graded warning mechanism based on comprehensive fault signals to provide graded warnings for equipment operation status; Design an adaptive adjustment mechanism to optimize the fault scoring model in the long term; After the real-time data of the campus hot water equipment is collected in real time, a state transition model is constructed to predict the current state according to the current sensor reading and the state history, and the predicted current state is used as the identified current state, specifically: Assume that the state at the previous moment is , define the state transition probability , and dynamically adjusted according to the sensor data, the state transition of the state transition model is expressed as follows: ； in, Indicates current data, Indicates water temperature data, represents water pressure data, t represents time t, Represents a fixed time interval, Indicates the current state. For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state Make estimates; The state transition rules are defined as follows: ； ； ； Among them, if the current Greater than zero and the water temperature rise rate Greater than the set threshold When the device is in heating state If the current Near zero and water temperature Keep stable, that is , the equipment is in the insulation state If the current If the temperature is zero and the water temperature gradually decreases, the device is in idle state. ; At the same time, a memory decay factor is introduced in the state transfer model , weighted calculation of state history is performed as follows: ； in, represents the transition probability based on historical state data, represents the transition probability calculated by the current sensor data, Control the balance of historical and current data.

2. A remotely monitored campus hot water equipment failure early warning method according to claim 1, characterized in that: The current working state of the device depends on the data output by the sensor: water temperature data, water pressure data and current data; The dynamically performing fault detection according to the current working state of the device and generating a fault detection signal specifically includes: In the heating state Perform fault detection under In the insulation state Perform fault detection under In the idle state Perform fault detection under The heating state , insulation state and idle state Comprehensive fault detection signal .

3. A remotely monitored campus hot water equipment failure early warning method according to claim 2, characterized in that: In the heating state Fault prediction is performed as follows: ； in, Indicates the heating efficiency, represents the rate of change of temperature, Indicates current data, is the energy loss coefficient, which is related to the equipment water pressure The fluctuation of If the heating efficiency Below the set minimum heating efficiency threshold , then the heating element has an inefficient fault; if the water pressure fluctuates greatly, that is Greater than a preset value , then increase value, conduct heating efficiency analysis; when abnormal heating efficiency is detected, a fault signal is generated , with heating element status information; In the insulation state The fault detection is performed as follows: ； in, is the water temperature change, if ,in is the threshold value of water temperature fluctuation allowed, If it is the insulation energy conservation term, it is judged that the equipment has poor insulation or abnormal heat dissipation, and a fault signal is generated ; In the idle state The fault detection is performed as follows: ； Among them, in the idle state, it is determined whether there is abnormal current leakage, that is, Or the water temperature does not drop normally, i.e. If there is current leakage or the water temperature drops slowly, it is judged as a fault and a signal is generated. ; The heating state , insulation state and idle state Comprehensive fault detection signal , which is expressed as follows: ； in, , , Represent the fault detection signals in heating, insulation and idle states respectively.

4. A remotely monitored campus hot water equipment failure early warning method according to claim 1, characterized in that: According to the fault detection signal, the energy conservation analysis based on thermodynamics is used to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems, and output a comprehensive fault signal, which is divided into heating state, insulation state and idle state.

5. A remotely monitored campus hot water equipment failure warning method according to any one of claims 1 to 4, characterized in that: The energy calculation in the heating state is expressed as follows: Heating status First, calculate the power input of the device. Based on the current and voltage readings of the device, the power input Calculated as: ； in, is the electrical energy input during heating, is the voltage, is the current value, is the time interval; Calculate the heat output during the heating process by changing the water temperature : ； in, is the quality of the water, is the specific heat capacity of water, Indicates the rate of change of water temperature ; Energy difference Reflects the degree of energy loss: ； if If the value is too large, it indicates that there is energy loss in the heating process, which is due to aging of the heating elements or reduced system energy efficiency; Introducing an innovative regularization term , which is used to compensate for the energy change caused by external pressure fluctuations, is expressed as follows: ； in, represents energy loss, is the adjustment factor, is the measured value of the current water pressure; if Exceeding a predetermined threshold , the system generates a heating energy loss fault signal , indicating that there is a serious energy loss during the heating process; In the insulation state Under temperature change Exceeding the set threshold , and no current input , there is energy leakage or abnormal heat dissipation: ； in, It is the temperature fluctuation threshold during the insulation stage. When the temperature fluctuation of the device exceeds this value and there is no additional energy input, it is judged that there is a fault in the insulation state of the device, and a insulation state loss fault signal is generated. ; In idle state The device should stop working and the current should be zero; if the current Greater than a certain threshold , it indicates that there is an electrical fault or current leakage in the device: ； At the same time, the water temperature drop rate is monitored in real time; if the water temperature does not drop significantly, , there is a poor heat dissipation problem, and a corresponding fault signal is generated .

6. A remotely monitored campus hot water equipment failure early warning method according to claim 5, characterized in that: The comprehensive fault signal is expressed as follows: ； in, Indicates a comprehensive fault signal.

7. A remotely monitored campus hot water equipment failure early warning method according to claim 5, characterized in that: The fault scoring model combines energy loss And the fault frequency are used to calculate the fault score, which is expressed as follows: ； in, Rate the fault, and are weighted coefficients, respectively controlling energy loss With fault signal , the impact on the ratings; It is the adjustment coefficient under different fault states to ensure that the influence weight of each state on the fault score is reasonable.

8. A remotely monitored campus hot water equipment failure early warning method according to claim 7, characterized in that: Introducing a warning regularization term into the fault scoring model , the effect of device status change in the corrected scoring model is expressed as follows: ； in, represents the corrected fault score, Used to adjust the warning sensitivity of the equipment under high load or low load conditions; The hierarchical early warning mechanism includes: Low-level warning: , indicating that the equipment is operating basically normally, with slight energy loss or occasional failures, and regular maintenance is recommended; Intermediate warning: , indicating that energy loss or fault signals frequently occur during equipment operation, which poses a potential risk. It is recommended to troubleshoot the problem in advance; Advanced warning: , indicating that a serious fault or long-term energy abnormality has occurred during the operation of the equipment, and the equipment must be shut down for inspection immediately to avoid further deterioration of the system.

9. A remotely monitored campus hot water equipment failure early warning method according to claim 8, characterized in that: The designed adaptive adjustment mechanism performs long-term optimization on the fault scoring model, specifically including: Adaptively adjust the energy loss weight according to the current corrected fault score of the device; Adjust the fault frequency weight according to the fault signals under different equipment states; Introduce feedback regularization term to compensate for the fault detection deviation caused by external disturbances; By analyzing the changing trend of fault scores, the fault scoring model is optimized in the long term; The energy loss weight, fault frequency weight and feedback regularization term are optimized in a closed loop by analyzing their changing trends.

10. A remote monitoring campus hot water equipment failure warning system, characterized in that: The system comprises: The equipment data collection module is used to deploy sensors to collect real-time data of campus hot water equipment and identify the current working status of the equipment; A fault signal initialization module is used to dynamically perform fault detection according to the current working status of the equipment and generate a fault detection signal; A fault signal generation module is used to monitor and confirm whether the equipment has efficiency loss or energy abnormality problems based on the fault detection signal using energy conservation analysis based on thermodynamics, and output a comprehensive fault signal; The fault warning module is used to design a fault scoring model and a graded warning mechanism based on comprehensive fault signals to provide graded warnings for the equipment operation status; System optimization module, which is used to design an adaptive adjustment mechanism to optimize the fault scoring model in the long term; After the real-time data of the campus hot water equipment is collected in real time, a state transition model is constructed to predict the current state according to the current sensor reading and the state history, and the predicted current state is used as the identified current state, specifically: Assume that the state at the previous moment is , define the state transition probability , and dynamically adjusted according to the sensor data, the state transition of the state transition model is expressed as follows: ； in, Indicates current data, Indicates water temperature data, represents water pressure data, t represents time t, Represents a fixed time interval, Indicates the current state. For the current state The posterior probability is calculated based on the rate of change of sensor data and the previous state Make estimates; The state transition rules are defined as follows: ； ； ； Among them, if the current Greater than zero and the water temperature rise rate Greater than the set threshold When the device is in heating state If the current Near zero and water temperature Keep stable, that is , the equipment is in the insulation state If the current If the temperature is zero and the water temperature gradually decreases, the device is in idle state. ; At the same time, a memory decay factor is introduced in the state transfer model , weighted calculation of state history is performed as follows: ； in, represents the transition probability based on historical state data, represents the transition probability calculated by the current sensor data, Control the balance of historical and current data.

Images