Safety monitoring method and device for ground source heat pump system

Abstract

The invention relates to the technical field of ground source heat pumps, and discloses a safety monitoring method and device of a ground source heat pump system, multi-dimensional real-time data such as the rotating speed of a circulating pump, the pipeline pressure and the environment temperature are collected through a multi-point sensor, and dynamic monitoring of the running state is achieved through noise filtering and time sequence analysis. When abnormity is detected, a comprehensive operation matrix is constructed in combination with historical operation records, and stress distribution of the underground pipeline is reversely calculated through the structured grid division technology. Based on a stress distribution sequence, a future stress evolution trend is analyzed and predicted by adopting linear regression, and a potential safety risk level is determined. And finally, dynamically adjusting the data acquisition frequency according to the risk level, and generating a self-adaptive safety monitoring strategy. Intelligent safety monitoring of the ground source heat pump system is achieved, potential risks can be found in time, prevention measures can be taken, and the operation safety and reliability of the system are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ground source heat pumps, and in particular to a safety monitoring method and device for a ground source heat pump system. Background Art

[0002] As an efficient and energy-saving heating and cooling technology, ground source heat pump system plays a vital role in building energy management. Its stable operation is directly related to energy utilization efficiency and environmental sustainable development.

[0003] Currently, most systems only deploy a small number of temperature and flow sensors at underground pipe inlets and outlets or at key nodes in the main engine. These sensors are unable to fully reflect the local conditions of underground heat exchangers, pipeline branches, or terminal equipment. Inexpensive sensors may not be able to quickly capture transient changes, leading to lags in control strategies. Without accurate capture of dynamic changes, it is difficult to determine whether the system is experiencing an abnormal state. When an anomaly occurs, relying solely on current data cannot fully reveal the root cause.

[0004] Therefore, the current system's dynamic perception of operating status is insufficient, and the prediction and tracing analysis of potential risks lack depth, making it difficult to promptly detect and respond to hidden problems in complex environments. Therefore, it is necessary to solve the key issues of safe operation of ground-source heat pump systems. Summary of the Invention

[0005] The present invention provides a safety monitoring method and system for a ground source heat pump system, so as to realize the safe operation of the ground source heat pump system.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a safety monitoring method for a ground source heat pump system, comprising:

[0007] Acquire a multidimensional data stream of the ground source heat pump system in real time, and filter the multidimensional data stream according to a preset noise threshold to obtain a real-time parameter set;

[0008] Determining a dynamic change trend based on the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, comparing the real-time parameter set with associated fields of a historical operation log to obtain an abnormal time window;

[0009] Extracting corresponding historical operation records according to the abnormal time window and generating a complete historical multidimensional data set, constructing a comprehensive operation matrix according to the complete historical multidimensional data set, and obtaining a stress distribution sequence according to the comprehensive operation matrix;

[0010] If the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, the stress evolution trend within a preset time period is predicted, the potential safety risk signal level is determined, the acquisition frequency is dynamically adjusted according to the potential safety risk signal level, and an adaptive safety monitoring strategy is generated.

[0011] Preferably, the real-time acquisition of the multi-dimensional data stream of the ground source heat pump system, filtering the multi-dimensional data stream according to the preset noise threshold to obtain a real-time parameter set, comprising:

[0012] Obtain the real-time data stream through the multi-point sensor, and eliminate the abnormal values exceeding the noise threshold to obtain a preliminary data set;

[0013] Smooth the preliminary data set, and if the environmental temperature fluctuation in a continuous time period is detected to exceed a preset temperature fluctuation threshold, an abnormal data segment corresponding to the environmental temperature fluctuation is obtained;

[0014] Analyze the trend of the abnormal data segment, and if the correlation between the circulating pump speed and the pipeline pressure peak deviates from the normal range, adjust the abnormal data to obtain an adjusted parameter set;

[0015] Reset the noise threshold, calibrate the adjusted parameter set, and determine whether it meets the requirements of the system running state, and if so, obtain the real-time parameter set.

[0016] Preferably, the determination of the dynamic change trend according to the real-time parameter set comprises:

[0017] Obtain the real-time data of the circulating pump from the real-time parameter set and perform smoothing processing to obtain smoothed speed sequence data and extract characteristic values of fluctuation periods therefrom;

[0018] Compare the characteristic values with a preset characteristic threshold, and if they exceed the threshold, obtain an abnormal fluctuation period and an abnormal segment;

[0019] Analyze the correlation between the abnormal fluctuation period and the dynamic change trend, dynamically evaluate the running state of the circulating pump according to the correlation, and determine whether it meets a preset standard, and if so, determine the dynamic change trend.

[0020] Preferably, if the dynamic change trend deviates from a preset normal range, compare the real-time parameter set with a correlation field of a historical running log to obtain an abnormal time window, comprising:

[0021] If the dynamic change trend deviates from the preset normal range, obtain the abnormal segment;

[0022] Obtain the corresponding field in the historical running log according to the abnormal segment and compare it to determine the abnormal time window;

[0023] Further comprising: extracting fluctuation features of the dynamic change trend according to the abnormal time window, if the fluctuation features continuously deviate from a preset fluctuation threshold, analyzing running state data in the time window to determine an abnormal specific distribution interval.

[0024] Preferably, the extracting corresponding historical running records according to the abnormal time window and generating a complete historical multi-dimensional data set comprises:

[0025] Obtaining historical running records of the abnormal time window to obtain an initial data set;

[0026] If it is found that there is data missing in the initial data set, the complete historical multi-dimensional data set is obtained by completing the processing through historical data.

[0027] Preferably, the constructing a comprehensive running matrix according to the complete historical multi-dimensional data set comprises:

[0028] Calibrating the accuracy of the complete historical multi-dimensional data set to obtain calibrated data and extract an influence factor of environmental temperature fluctuation to generate a fluctuation feature set containing the influence factor;

[0029] Generating a comprehensive running matrix containing time sequence features according to the fluctuation feature set.

[0030] Preferably, the obtaining a stress distribution sequence according to the comprehensive running matrix comprises:

[0031] Obtaining historical running data of an underground pipeline, and structuring and layering the historical running data to obtain a layered data set;

[0032] Integrating a spatial mapping relationship of the historical running data according to the layered data set to generate a running matrix;

[0033] Extracting a characteristic value set related to stress distribution from the running matrix, and correcting the characteristic value set to obtain stress distribution data;

[0034] If the stress distribution data deviates from a preset distribution threshold, the stress distribution data is grouped to generate the stress distribution sequence.

[0035] Preferably, if a stress distribution peak point in the stress distribution sequence exceeds a safety threshold, a stress evolution trend in a preset time period is predicted, and a potential safety risk signal level is determined, comprising:

[0036] Obtaining historical stress records of an underground pipeline, and extracting a stress change set;

[0037] Obtaining a stress change rate according to the stress change set, obtaining a stress change map according to the stress change rate and the historical stress record, and determining a prediction coefficient combination;

[0038] If the coefficients in the prediction coefficient combination exceed a preset coefficient threshold, a stress evolution sequence is generated according to the prediction time span to obtain an evolution trend set;

[0039] The evolution trend set is compared with the safety threshold to generate a safety risk sequence and determine the potential safety risk signal level.

[0040] Preferably, dynamically adjusting the acquisition frequency according to the potential safety risk signal level to generate an adaptive safety monitoring strategy includes:

[0041] Obtaining an acquisition frequency reference value according to the potential safety risk signal level, dynamically adjusting the acquisition frequency according to the acquisition frequency reference value, generating a frequency adjustment plan, and determining an adjusted frequency range;

[0042] If the adjusted frequency range exceeds the preset frequency threshold, a speed optimization sequence is generated in combination with the historical records of the circulating pump to obtain an optimized speed range;

[0043] Dynamically matching the monitoring period of the pipeline pressure peak with the optimized speed range to generate a monitoring period plan, and determining whether the monitoring period plan meets the safety monitoring requirements;

[0044] If so, the adjusted frequency range, the optimized speed range and the monitoring period scheme are integrated to generate the adaptive safety monitoring strategy.

[0045] In a second aspect, the present invention provides a safety monitoring device for a ground source heat pump system, comprising:

[0046] A first acquisition module is used to acquire a multi-dimensional data stream of the ground source heat pump system in real time, and obtain a real-time parameter set after filtering the multi-dimensional data stream according to a preset noise threshold;

[0047] A second acquisition module is configured to determine a dynamic change trend based on the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, compare the real-time parameter set with associated fields in a historical operation log to obtain an abnormal time window;

[0048] a construction module, configured to extract corresponding historical operation records according to the abnormal time window and generate a complete historical multidimensional data set, construct a comprehensive operation matrix according to the complete historical multidimensional data set, and obtain a stress distribution sequence according to the comprehensive operation matrix;

[0049] The generating module is used for predicting a stress evolution trend in a preset time period, determining a potential safety risk signal level, dynamically adjusting a collection frequency according to the potential safety risk signal level, and generating an adaptive safety monitoring strategy if a stress distribution peak point in the stress distribution sequence exceeds a safety threshold.

[0050] Compared with the prior art, the present application has the following beneficial effects:

[0051] (1) The present application obtains multi-dimensional data flow of the ground source heat pump system in real time, filters the multi-dimensional data flow according to a preset noise threshold to obtain a real-time parameter set, and this method considers efficiency and accuracy in real-time monitoring, thereby providing important support for equipment management in an industrial scene.

[0052] (2) The present application determines a dynamic change trend according to the real-time parameter set, compares the real-time parameter set and a historical operation log in a correlation field if the dynamic change trend deviates from a preset normal range, obtains an abnormal time window, improves the reliability of data analysis, and provides strong support for stable operation of the ground source heat pump system.

[0053] (3) The present application extracts corresponding historical operation records according to the abnormal time window and generates a complete historical multi-dimensional data set, constructs a comprehensive operation matrix according to the complete historical multi-dimensional data set, obtains a stress distribution sequence according to the comprehensive operation matrix, and uses a multi-level analysis method to effectively improve comprehensive mastery of the pipeline operation state and reduce potential risks caused by stress abnormalities.

[0054] (4) In the present application, if a stress distribution peak point in the stress distribution sequence exceeds a safety threshold, a stress evolution trend in a preset time period is predicted, a potential safety risk signal level is determined, a collection frequency is dynamically adjusted according to the potential safety risk signal level, and an adaptive safety monitoring strategy is generated. This ensures multi-dimensional safety prevention and control and constructs a rigorous safety monitoring network. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a safety monitoring method flow chart of a ground source heat pump system provided by an embodiment of the present application;

[0056] Figure 2 is a safety monitoring module structure diagram of a ground source heat pump system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0057] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0058] With reference to Figure 1 The embodiment of the present application provides a safety monitoring method flow chart of a ground source heat pump system, comprising the following steps:

[0059] S1, real-time acquisition of multi-dimensional data stream of the ground source heat pump system, filtering of the multi-dimensional data stream according to a preset noise threshold to obtain a real-time parameter set;

[0060] S2, determination of a dynamic change trend according to the real-time parameter set, comparison of the real-time parameter set with a historical operation log in a correlation field if the dynamic change trend deviates from a preset normal range to obtain an abnormal time window;

[0061] S3, extraction of corresponding historical operation records according to the abnormal time window and generation of a complete historical multi-dimensional data set, construction of a comprehensive operation matrix according to the complete historical multi-dimensional data set, and obtaining of a stress distribution sequence according to the comprehensive operation matrix;

[0062] S4, prediction of a stress evolution trend in a preset time period if a stress distribution peak point in the stress distribution sequence exceeds a safety threshold, determination of a potential safety risk signal level, dynamic adjustment of a collection frequency according to the potential safety risk signal level, and generation of an adaptive safety monitoring strategy.

[0063] In step S1, real-time acquisition of multi-dimensional data stream of the ground source heat pump system, filtering of the multi-dimensional data stream according to a preset noise threshold to obtain a real-time parameter set.

[0064] Among them, in the ground source heat pump system, the multi-dimensional data stream (Multi-dimensional Data Stream) refers to a multi-source, multi-type, multi-dimensional dynamic data set collected in real time during system operation. These data are continuously generated through time series or event-driven methods, and are used for monitoring, analysis, optimization of system performance or implementation of intelligent control. The core purpose is to comprehensively reflect the system state by fusing information of different dimensions, and to support efficient decision-making (such as energy efficiency optimization, fault warning, etc.).

[0065] Preferably, the real-time acquisition of multi-dimensional data stream of the ground source heat pump system, the filtering of the multi-dimensional data stream according to a preset noise threshold to obtain a real-time parameter set, comprises:

[0066] A real-time data stream is acquired by a multi-point sensor, outliers exceeding the noise threshold are removed, and a preliminary data set is obtained;

[0067] The preliminary data set is smoothed, and if the environmental temperature fluctuation in a continuous time period exceeds a preset temperature fluctuation threshold, the corresponding abnormal data segment is acquired;

[0068] The trend of the abnormal data segment is analyzed, and if the correlation between the circulating pump speed and the pipeline pressure peak deviates from the normal range, the abnormal data is adjusted, and an adjusted parameter set is obtained;

[0069] The noise threshold is reset, the adjusted parameter set is calibrated, and it is determined whether it meets the requirements of the system operating state. If so, the real-time parameter set is obtained.

[0070] In a ground source heat pump system (GSHP), the correlation between the circulating pump speed and the pipeline pressure peak reflects the dynamic relationship between the system hydraulic characteristics and the operation of the power equipment, directly affecting the system energy efficiency, stability and equipment life.

[0071] For example, in the real-time monitoring scenario of circulating pump operation, the processing of multi-point sensor data stream first needs to preliminarily screen key indicators such as circulating pump speed, pipeline pressure peak and environmental temperature fluctuation. Assuming that the reasonable range of speed is 800-1200 revolutions per minute (RPM), the range of pipeline pressure peak is 0.4-1.2 megapascals, and the range of environmental temperature fluctuation is -5 to 5 degrees Celsius. By presetting threshold rules, outliers with a speed of 1500 revolutions per minute or a pressure peak of 1.3 megapascals are removed, and a preliminary cleaned data set is obtained. This process can effectively reduce noise interference in subsequent analysis and improve data quality.

[0072] In the secondary processing stage, the data is smoothed in combination with data acquisition frequency such as once per second and data stream stability. Assuming that the environmental temperature fluctuates more than 5 degrees Celsius in 10 consecutive minutes, from 20 degrees Celsius to 28 degrees Celsius, the data in this time period is marked as an abnormal data segment. Such marking helps to quickly locate potential problems and ensure the pertinence of subsequent analysis.

[0073] For abnormal data segments, parameters are detected in combination with system operating state and historical record analysis trends. Assuming that the sliding window is set to 5 minutes, it is found that when the circulating pump speed is 900 revolutions per minute, the pipeline pressure peak is 1.6 megapascals, deviating from the normal correlation range, i.e. the speed and pressure should show a positive correlation trend. At this time, the data is adjusted, and the calibrated pressure value is 1 megapascal, and the calibrated parameter set is obtained. Such adjustment can avoid misjudgment caused by single indicator abnormality and improve system stability.

[0074] In the final refining stage, the calibrated parameter set is integrated, and all indicators are comprehensively processed. Assuming that the fusion rotation speed is 950 rpm, the pressure peak is 1.2 MPa, and the temperature fluctuation is 3°C, the preset threshold rule is applied again to verify that the data meets the system operation state requirements, and the final refining result is obtained. This process ensures the comprehensiveness and reliability of the data, providing a solid foundation for subsequent decision-making.

[0075] For example, from the technical effect, the above multi-stage processing method can significantly improve data accuracy and reduce false positive rate. For example, in the case of environmental temperature fluctuations, marking abnormal segments and adjusting in combination with historical data can effectively avoid system misjudgment caused by short-term interference and ensure the safety of the circulating pump operation. At the same time, it also makes the system more adaptable to complex working conditions, especially when the correlation between pressure peak and rotation speed deviates, the calibration process can correct the deviation in time and prolong the service life of the equipment. Overall, this method balances efficiency and accuracy in real-time monitoring, providing important support for equipment management in industrial scenarios.

[0076] In step S2, a dynamic change trend is determined according to the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, the real-time parameter set is compared with the associated fields of the historical operation log to obtain an abnormal time window.

[0077] In a ground source heat pump system (GSHP), dynamic change trend and abnormal time window detection are core links of system intelligent monitoring and fault diagnosis. The dynamic change trend refers to the regular pattern of system operating parameters changing with time, which is usually captured by real-time sensor data, reflecting the development process of system state, energy efficiency or potential faults. Through multi-parameter correlation analysis (such as rotation speed + pressure + temperature), the abnormal time window (i.e. the time period of abnormality) is identified.

[0078] Preferably, the determination of the dynamic change trend according to the real-time parameter set comprises:

[0079] Real-time data of the circulating pump is obtained from the real-time parameter set and smoothed to obtain smoothed rotation speed sequence data and extract characteristic values of fluctuation periods therefrom;

[0080] The characteristic values are compared with preset characteristic threshold values, and if they exceed, abnormal fluctuation periods and abnormal segments are obtained;

[0081] The correlation degree between the abnormal fluctuation periods and the dynamic change trend is analyzed, and the running state of the circulating pump is dynamically evaluated according to the correlation degree to determine whether it meets the preset standard, and if so, the dynamic change trend is determined.

[0082] In the ground source heat pump system, the correlation between abnormal fluctuation period and dynamic trend reflects the statistical or causal relationship between the abnormal time characteristics (such as periodicity, suddenness) and the overall running trend (such as long-term performance degradation, load change) when the system fails or runs in a non-steady state. This correlation analysis is the core of fault diagnosis and health management. At the same time, dynamic evaluation of the circulating pump running state through the correlation between abnormal fluctuation period and dynamic trend needs to combine time series analysis, multi-scale feature extraction and physical model verification to form a closed-loop diagnosis strategy.

[0083] For example, in the ground source heat pump system, real-time data processing for circulating pump speed, smooth operation is the key. By reducing short-term noise in the data, the main trend is retained, which helps to more accurately analyze the speed change. Assuming that the collected speed data is recorded every second, the original data fluctuates between 900-1100 revolutions per minute, but occasionally there are short peak values, such as 1200 revolutions per minute. Using the moving average smoothing method, a 5-second window is used to calculate the average speed every 5 seconds, and the smoothed sequence data is obtained, such as 950 revolutions per minute. This processing can effectively eliminate sensor noise interference and provide a reliable basis for subsequent analysis.

[0084] Based on the smoothed speed sequence data, extract the fluctuation period characteristic value. Decomposition divides the data into trend, periodic and residual parts, focusing on periodic fluctuations.

[0085] Analysis found that the speed data has a periodic fluctuation every 30 minutes, with a characteristic value of speed change amplitude of about 50 revolutions per minute. If the preset threshold is that the fluctuation amplitude does not exceed 30 revolutions per minute, this period is marked as an abnormal fluctuation period, and the related data segment is recorded as an abnormal segment. This marking helps to quickly locate potential unstable operation problems.

[0086] For abnormal segment data, analyze the correlation between fluctuation period and dynamic trend combined with historical records. Assuming that the sliding window is set to 10 minutes, it is found that the speed in a certain abnormal segment periodically rises to 1050 revolutions per minute, and the historical records show that the pipeline pressure is stable during this period. It is speculated that the speed anomaly may be caused by external environmental changes. By comparing historical data, it is confirmed that under normal conditions, speed and pressure should change synchronously, and the corresponding relationship data between fluctuation period and stable pressure is obtained. This analysis can reveal the potential causes of abnormalities.

[0087] Combined with the constraint conditions of environmental influence, such as environmental temperature range of 15-25 degrees Celsius, the running state of the speed interval is dynamically evaluated.

[0088] For example, if the ambient temperature rises to 28°C during a certain period and the speed fluctuates by 60 rpm, exceeding the preset standard, the assessment results will indicate that this speed range does not meet the requirements and is marked as an abnormal trend. Conversely, if the temperature remains stable at 20°C and the speed fluctuates by 25 rpm, the assessment is normal. This dynamic assessment can promptly reflect changes in system operating status.

[0089] It can be understood that the above multi-level processing method ensures the comprehensiveness and accuracy of the speed data through smoothing, decomposition, sliding window analysis and conditional judgment, layer by layer.

[0090] Smoothing reduces noise, extracts periodic features, reveals dynamic correlations, and provides a final assessment of operational status. These steps mutually support each other, improving the reliability of data analysis and providing strong support for the stable operation of the ground-source heat pump system.

[0091] Preferably, if the dynamic change trend deviates from a preset normal range, comparing the real-time parameter set with the associated fields of the historical operation log to obtain the abnormal time window includes:

[0092] If the dynamic change trend deviates from the preset normal range, obtaining the abnormal segment;

[0093] Obtaining corresponding fields in the historical operation log according to the abnormal fragment and performing comparison to determine the abnormal time window;

[0094] It then includes: extracting the fluctuation characteristics of the dynamic change trend according to the abnormal time window, and if the fluctuation characteristics continue to deviate from the preset fluctuation threshold, analyzing the operating status data in the time window to determine the specific distribution interval of the abnormality.

[0095] In one possible implementation, for the operating status data of the circulation pump in the ground source heat pump system, the acquisition and comparison of the real-time parameter set is key.

[0096] For example, a sensor collects operating current data from a circulation pump every second. Assume the normal threshold range is 80-660 amperes (A). If the current reaches 680A within a certain period, exceeding the preset threshold, this data segment will be marked as abnormal. By comparing real-time data with the threshold, deviations in operating status can be quickly identified.

[0097] Preferably, the timestamp and specific parameter values ​​of the abnormality are recorded, such as an abnormality occurring between 10:00 and 10:05, with a current value of 680 A. This marking provides an accurate data basis for subsequent analysis.

[0098] Based on the abnormal flag data, corresponding fields such as pipeline flow and ambient temperature are extracted from historical logs. By comparing real-time current with flow data in historical logs, it is found that the flow stabilizes at 1189 liters / hour during the abnormal period, while under normal circumstances the current and flow are positively correlated. Assuming that historical data shows that when the normal current is 90A, the flow is about 1189 liters / hour, and the flow in the abnormal period does not increase synchronously. It is inferred that the abnormality may be related to internal resistance of the pump or external interference. By timestamp alignment, it is determined that the abnormal time window is 10:00-10:10, providing a clear range for subsequent analysis.

[0099] In one possible implementation, for time window data, the fluctuation characteristics of dynamic trends are extracted.

[0100] For example, with a 5-minute window, the current data in 10:00-10:10 is analyzed, and it is found that the current value fluctuates between 550-660A, and the fluctuation period is about 2 minutes, deviating from the normal fluctuation amplitude threshold of 0.5 ampere. By segment-by-segment analysis, the abnormal high-frequency change of current is captured. Combined with the state evaluation condition, if the fluctuation amplitude continues to exceed 0.5 ampere, it is judged as a trend deviation. The evaluation result shows that the system is unstable during this time window, and there may be mechanical wear or external power fluctuation.

[0101] For the evaluation result of trend deviation, parameter comparison further analyzes the running state data in the time window to determine the specific distribution interval of the anomaly.

[0102] For example, analysis finds that the current value from 10:02 to 10:04 is continuously higher than 550A, accompanied by a slight increase in motor temperature to 45 degrees Celsius, while the normal temperature is 40 degrees Celsius. Parameter comparison confirms that the anomaly is mainly concentrated in the superimposed interval of high current and high temperature through multi-dimensional data comparison.

[0103] Preferably, historical data will also be combined to verify whether similar anomalies have repeatedly occurred in this interval.

[0104] For example, historical records show that current anomalies also occurred during the same time period last week, suggesting that they may be related to the daily peak electricity consumption period. Such analysis helps to accurately locate the cause of the anomaly and provides data support for system maintenance.

[0105] For example, for constraint conditions such as ambient temperature 20-25 degrees Celsius, state evaluation will further verify whether the anomaly is affected by external factors. Assuming that the ambient temperature during the abnormal period is 26 degrees Celsius, which exceeds the normal range, it may cause the heat dissipation efficiency of the motor to decrease, thereby causing current anomalies. By comparing the correlation between temperature and current, it is confirmed that the abnormal distribution interval is highly coincident with the high temperature period. This multi-level analysis, from real-time data to historical comparison, to environmental constraints, layer by layer, ensures the comprehensiveness of anomaly detection.

[0106] In step S3, the corresponding historical operation records are extracted according to the abnormal time window and a complete historical multidimensional data set is generated. A comprehensive operation matrix is ​​constructed according to the complete historical multidimensional data set, and a stress distribution sequence is obtained according to the comprehensive operation matrix.

[0107] Among them, in the ground source heat pump system (GSHP), the integrated operation matrix and stress distribution sequence are two key data modeling methods used for system performance analysis, fault diagnosis and life prediction. The integrated operation matrix (Integrated Operation Matrix) is a multi-dimensional data structure that integrates the time series or state data of various dynamic parameters (such as temperature, pressure, flow, energy consumption, etc.) of the ground source heat pump system during operation to form a unified mathematical representation. Its essence is a characteristic matrix used to describe the operating status of the system in multiple dimensions and time scales. The stress distribution sequence (Stress Distribution Sequence) is a sequence data that describes the mechanical or thermal stress of key components (such as circulation pumps, heat exchangers, and pipelines) in the ground source heat pump system over time. It is calculated through sensors or physical models to reflect the fatigue accumulation and potential damage risk of components.

[0108] Preferably, extracting corresponding historical operation records according to the abnormal time window and generating a complete historical multidimensional data set includes:

[0109] Obtaining historical operation records of the abnormal time window to obtain an initial data set;

[0110] If data is found to be missing in the initial data set, the data is supplemented using historical data to obtain the complete historical multidimensional data set.

[0111] For example, in a ground-source heat pump system, detecting anomalies in the operating status of the circulating pump and obtaining historical operating records corresponding to the abnormal time window from a pre-established database is a key step in analyzing the root cause of the anomaly. Assuming the abnormal time window is 10:00-10:10, the circulating pump speed range, ambient temperature fluctuation range, and pipeline pressure peak data within this time period are filtered out to form an initial data set. Specifically, speed data such as 800-1000 revolutions per minute, ambient temperature fluctuations such as 22-24 degrees Celsius, and pipeline pressure peaks such as 1.6 MPa are extracted to form a multi-dimensional data set, laying the foundation for subsequent analysis.

[0112] During data cleaning of the initial dataset, we check for null values ​​or missing records. For example, if the speed data at 10:05 is empty, we compare it with historical data and find the average speed for the same time period over the past week, such as 850 rpm, to complete the data set and ensure its integrity. This approach effectively avoids analytical bias caused by missing data and provides reliable data support for subsequent multidimensional analysis.

[0113] Preferably, said constructing a comprehensive operation matrix based on said complete historical multidimensional data set comprises:

[0114] Performing precision calibration on the complete historical multidimensional data set to obtain calibrated data and extracting influencing factors of ambient temperature fluctuations to generate a fluctuation feature set including the influencing factors;

[0115] A comprehensive operation matrix including time series characteristics is generated according to the fluctuation characteristic set.

[0116] Among them, in the ground source heat pump system (GSHP), the fluctuation feature set and the integrated operation matrix are two key data modeling methods, which are used to comprehensively characterize the dynamic behavior of the system, anomaly detection and energy efficiency optimization. The fluctuation feature set (Fluctuation Feature Set) refers to a set of features extracted from the system operation data that can reflect the multi-scale abnormal fluctuation pattern, including quantitative indicators of multiple dimensions such as time, frequency, and statistics. Its core purpose is to identify potential faults or performance degradations by mathematically characterizing the dynamic change laws of system parameters. The integrated operation matrix (Integrated Operation Matrix) is a multi-dimensional matrix of time and space fusion, which integrates the time series data of all key parameters of the system and their derived features to form a structured representation. Its essence is the extension of the fluctuation feature set in the time dimension, which is used to support the input or visual analysis of machine learning models.

[0117] The fluctuation feature set is the "fingerprint" of the system's dynamic behavior, revealing abnormal patterns through multi-domain features. The comprehensive operation matrix is ​​a spatiotemporal integration of features, providing structured input for data-driven models. The combination of the two can achieve closed-loop management from "data" to "decision-making", improving the reliability and energy efficiency of the ground-source heat pump system.

[0118] For example, in monitoring the operation of a ground-source heat pump system, obtaining raw data on sensor accuracy and ambient temperature is a critical first step. Sensor accuracy directly impacts data reliability. For example, if a sensor's accuracy is ±0.5%, and the collected ambient temperature data is within the range of 20-25°C, the raw data may exhibit slight deviations. This data can be corrected using a historical calibration model. For example, adjusting a temperature value collected at a specific point in time from 23.2°C to 23.0°C to approximate the actual value, ensuring the accuracy of subsequent analysis.

[0119] For example, extracting the influencing factors of ambient temperature fluctuations from calibrated data and ambient temperature data can help identify the specific impact of temperature changes on system operation. For example, if the temperature rapidly rises from 22°C to 24°C over a certain period of time, factor analysis may reveal the correlation between temperature fluctuations and the operational stability of the circulating pump, generating a fluctuation feature set containing the influencing factors. This feature set can provide important reference for subsequent analysis, especially when determining whether the temperature change is the cause of the anomaly.

[0120] By integrating the acquisition frequency and matrix dimensions, an operation matrix containing time series characteristics can be generated. Assuming an acquisition frequency of once per minute, and matrix dimensions covering multiple metrics such as temperature and rotational speed, the resulting operation matrix can intuitively reflect the system's operating status within a specific time period. For example, within the time window of 10:00-10:10, the matrix might show the corresponding relationship between temperature fluctuations and rotational speed changes, laying the foundation for further analysis.

[0121] If the service distribution of the operating matrix deviates from the preset threshold, such as temperature fluctuations exceeding 2 degrees Celsius or speed variations outside the normal range, cluster analysis is performed to process the matrix dimensions and attribute associations to identify the abnormal service distribution characteristics. If the analysis reveals an abnormal clustering of temperature fluctuations and speed changes within a certain time period, it may be due to a sudden change in ambient temperature causing an abnormal system response. The clustering results can clarify the specific characteristics of the abnormal distribution. This approach helps quickly locate the root cause of the problem and provides a basis for system adjustments.

[0122] From a business perspective, the impact of ambient temperature fluctuations on equipment operation in a ground-source heat pump system cannot be ignored. The aforementioned series of analyses forms a complete analytical chain, from raw data calibration to feature extraction, matrix construction, and anomaly distribution identification. Especially in scenarios with large temperature fluctuations, such as sudden drops in ambient temperature in winter, system operating parameters may deviate from their normal range, making timely identification and adjustment of parameters particularly important. This multi-dimensional analysis approach can effectively improve system operational stability and mitigate potential risks caused by environmental factors.

[0123] As an extension, if a factor is found to be consistently high over a specific time period—for example, if the temperature fluctuation factor exceeds the threshold for three consecutive days—further analysis can be conducted to determine its correlation with other system parameters. This analysis helps to move beyond the core issue to its potential impacts, comprehensively covering potential risk points and supporting the development of long-term maintenance strategies.

[0124] Preferably, obtaining the stress distribution sequence according to the comprehensive operation matrix includes:

[0125] Acquiring historical operation data of underground pipelines, and performing structured hierarchical organization on the historical operation data to obtain a hierarchical data set;

[0126] Integrate the spatial mapping relationship of the historical operation data according to the hierarchical data set to generate an operation matrix;

[0127] extracting an eigenvalue set related to stress distribution from the operation matrix, and correcting the eigenvalue set to obtain stress distribution data;

[0128] If the stress distribution data deviates from a preset distribution threshold, the stress distribution data is grouped to generate the stress distribution sequence.

[0129] In a ground-source heat pump (GSHP) system, stress distribution sequences refer to dynamic data sequences of the mechanical and thermal stresses experienced by key system components (such as circulation pumps, heat exchangers, and pipelines) during operation, along with their temporal variations. These sequences, obtained through sensor monitoring or physical model calculations, are used to quantify component fatigue damage, predict lifespan, and optimize operational strategies.

[0130] For example, in the field of underground pipeline operation monitoring, acquiring historical operating data is fundamental. This data typically includes information such as pipeline pressure, flow rate, and operating time. Using gridding tools to structure and layer this data allows complex operating data to be segmented by geographic location or time period, forming a primary grid dataset. For example, the area covered by a pipeline section is divided into 10x10 grid cells, with each cell corresponding to specific operating data for that section of the pipeline. For example, if the average pressure recorded within a cell is 1.1 MPa, this facilitates subsequent spatial analysis.

[0131] For example, a matrix can be used to integrate historical operational data with spatial mapping to generate a first operational matrix. This matrix can intuitively reflect the distribution of pipeline status within different grid cells. Assuming one column in the matrix represents the pressure value of a specific pipeline section, and another column represents the corresponding geographic coordinates, this method can identify key points in the pipeline's status. For example, if the pressure value of a grid cell is consistently high, it may indicate a potential risk area.

[0132] For example, when processing the operational matrix, we extract eigenvalue sets related to stress distribution. This process primarily uses historical data to infer the stress conditions likely experienced by the pipeline at different locations. For example, if the eigenvalues ​​for a particular pipeline segment indicate that stress is concentrated near a connection point, these eigenvalues ​​are corrected to obtain the initial stress distribution data, ensuring data accuracy. For example, after correction, the stress value at a certain point is adjusted from an initial 1.3 MPa to 1.1 MPa, which is closer to the actual state.

[0133] For example, if the stress distribution data deviates from the preset threshold, such as the stress value of a certain section of the pipeline exceeding the safe range of 1.6 megapascals, the data is grouped for processing. Clustering can help identify specific distribution patterns of stress anomalies, generate stress distribution sequences, and determine peak point distribution. Suppose the analysis finds that the stress values in a certain section of the pipeline in three consecutive grid cells are all at a peak, which may indicate that there is a structural problem in this area and further attention is needed.

[0134] From a business perspective, the stress distribution of underground pipelines is directly related to the long-term stability of the system. Through grid division and matrix construction, the spatial distribution characteristics of pipeline status can be clearly sorted out, and reverse calculation and data correction can help accurately locate stress abnormal points. Clustering and sequence generation further refine the feature recognition of abnormal distribution. This multi-level analysis method can effectively improve the overall understanding of the running status of the pipeline and reduce the potential risks caused by stress anomalies.

[0135] As an extended solution, if the stress distribution sequence of a certain section of the pipeline is found to be consistently high within a certain time period, such as the peak points being concentrated in the same area for a week, it can be further analyzed in conjunction with environmental factors such as ground subsidence or temperature changes. This approach helps extend from the core problem to potential impacts, covering risk points comprehensively, and providing support for developing long-term maintenance strategies.

[0136] In step S4, if the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, the stress evolution trend within a preset time period is predicted, the potential safety risk signal level is determined, the acquisition frequency is dynamically adjusted according to the potential safety risk signal level, and an adaptive safety monitoring strategy is generated.

[0137] Preferably, if the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, the stress evolution trend within a preset time period is predicted, the potential safety risk signal level is determined, including:

[0138] Obtain the historical stress record of the underground pipeline, and extract the stress change set;

[0139] Obtain the stress change rate according to the stress change set, obtain the stress change mapping according to the stress change rate and the historical stress record, and determine the prediction coefficient combination;

[0140] If the coefficient in the prediction coefficient combination exceeds the preset coefficient threshold, generate a stress evolution sequence according to the prediction time span, and obtain an evolution trend set;

[0141] Compare the evolution trend set with the safety threshold to generate a safety risk sequence, and determine the potential safety risk signal level.

[0142] Among them, in the ground source heat pump system (GSHP), the potential safety risk signal level, evolution trend collection and safety risk sequence are the core concepts used for system safety risk assessment and predictive maintenance. They realize the quantification, dynamic tracking and early warning of risks through multi-dimensional data analysis. The potential safety risk signal level (Potential Safety RiskSignal Level) is an indicator for grading the severity of abnormal signals (such as pressure exceeding the limit, temperature sudden change, etc.) detected in real time in the system. The risk level is usually divided into historical data and expert rules. The evolution trend collection (EvolutionTrend Collection) is a collection of trend models that reflects the dynamic development law of safety risks. It is used to predict the transformation path of risks from low levels to high levels. The safety risk sequence (Safety Risk Sequence) is a chain of risk events arranged in chronological order, recording the entire process from risk generation, development to final consequences, and is used for root cause analysis and responsibility tracing.

[0143] For example, in the field of underground pipeline operation monitoring, obtaining historical stress records and operating time records is fundamental to analysis. This data typically covers pipeline stress changes over different time periods and the corresponding operating time. Extracting the first set of data related to stress changes can help focus on key information.

[0144] For example, suppose that in the records of a certain section of pipeline over the past year, the stress value fluctuates significantly at a specific time point. These fluctuation data are extracted to form the first stress change set, laying the foundation for subsequent analysis.

[0145] For example, by analyzing the correlation between the stress change rate and the amount of historical data for a set of stress changes, a stress change map can be generated, which can then determine a combination of prediction coefficients. This process primarily explores whether the stress change trend exhibits a linear relationship with operating time or the amount of data accumulated. For example, if the stress change rate of a certain pipeline section exhibits a steady upward trend over a period of time with a large amount of data, the generated map might show a prediction coefficient of 0.8, exceeding the preset threshold of 0.5. This indicates a potential anomaly that requires further attention.

[0146] For example, when a coefficient in a combination of predicted coefficients exceeds a threshold, time series analysis is used to generate a stress evolution sequence based on the forecast time span, resulting in a set of evolution trends. This is primarily used to predict stress variations over a period of time. Assuming a forecast span of three months, the analysis reveals that the stress value of a particular pipeline section may peak at a specific time, for example, 1.1 MPa, approaching the upper safety limit. This provides important evidence for subsequent risk assessments.

[0147] For example, risk assessments are used to compare evolving trends with safety thresholds to generate a safety risk sequence and determine the level of potential safety risk signals. This process aims to identify whether stress changes pose a threat to pipeline operations. For example, if the safety threshold is 1.2 MPa and a pipeline section's stress values ​​in the predicted sequence are close to this value, the risk assessment might flag it as a medium-level risk signal, indicating the need for enhanced monitoring or preventive measures. This approach helps identify potential problems early and ensures the long-term stable operation of the pipeline system.

[0148] From a business perspective, stress analysis of underground pipelines is directly related to operational safety. Data screening and linear regression can reveal underlying patterns in stress variation, while time series analysis provides forecasts for future trends. Risk assessment further clarifies the severity of potential safety hazards. This multi-layered approach effectively enhances visibility into pipeline status, mitigates potential issues caused by stress anomalies, and provides a reliable basis for maintenance decisions.

[0149] As an extension, if the safety risk sequence indicates a consistently high risk level for a particular pipeline section over a specific time period, correlation analysis can be conducted in conjunction with other historical records, such as changes in the operating environment or external loads. For example, if the high-risk period coincides with surface construction, this may indicate the impact of external factors on pipeline stress. This comprehensive analysis helps to identify underlying causes, moving beyond the core issue and providing additional insights for developing targeted maintenance strategies.

[0150] Preferably, dynamically adjusting the acquisition frequency according to the potential safety risk signal level to generate an adaptive safety monitoring strategy includes:

[0151] Obtaining an acquisition frequency reference value according to the potential safety risk signal level, dynamically adjusting the acquisition frequency according to the acquisition frequency reference value, generating a frequency adjustment plan, and determining an adjusted frequency range;

[0152] If the adjusted frequency range exceeds the preset frequency threshold, a speed optimization sequence is generated in combination with the historical records of the circulating pump to obtain an optimized speed range;

[0153] Dynamically matching the monitoring period of the pipeline pressure peak with the optimized speed range to generate a monitoring period plan, and determining whether the monitoring period plan meets the safety monitoring requirements;

[0154] If so, the adjusted frequency range, the optimized speed range and the monitoring period scheme are integrated to generate the adaptive safety monitoring strategy.

[0155] Among them, in the ground source heat pump system (GSHP), the sampling frequency benchmark and rate optimization sequence are key technical concepts related to data monitoring and system performance optimization. The sampling frequency benchmark refers to the standard time interval at which the sensor or controller collects data on key parameters (such as temperature, pressure, flow, etc.) when the system is running. It is usually set based on physical process characteristics and control requirements. The rate optimization sequence is a dynamic adjustment sequence for the control parameters of the ground source heat pump system (such as the speed of the circulating pump and the valve opening). It optimizes the operating rate in real time through an algorithm to achieve the optimal balance between energy efficiency, stability or life.

[0156] The acquisition frequency baseline value represents the system's "temporal resolution" and determines data granularity. The speed optimization sequence serves as the control's "action guide," improving dynamic performance. Together, these two elements form the ground-source heat pump system's intelligent closed-loop "perception-decision-execution," a key technology for improving reliability, energy efficiency, and lifespan.

[0157] Specifically, in the process of building an adaptive security monitoring strategy, the system first automatically collects potential security risk signals and dynamically adjusts the collection frequency of real-time data streams according to the signal level.

[0158] For example, the system categorizes risk signals into low, medium, and high levels. When abnormal pipeline pressure fluctuations exceeding a preset threshold of 1.2 MPa are detected, the risk level is raised to high, and the data collection frequency is automatically adjusted from once per minute to once every 10 seconds. A sliding average calculation based on a time window is used to analyze the standard deviation of pressure data over the last five minutes. If the standard deviation is greater than 2.0, the trigger frequency is increased to ensure that abnormal trends are captured promptly.

[0159] Afterwards, by capturing abnormal trends and combining them with historical records of the circulating pump, a speed optimization sequence is generated and the optimized speed range is obtained. For example, key features are extracted from abnormal trends and a rule base is generated based on historical optimal records (such as "When ΔT < 2°C, reduce the speed by 5%"). Through historical data regression analysis, the optimal speed range under different operating conditions is found.

[0160] Next, the system optimizes the circulation pump speed range based on the collected data. Assuming that the normal speed range is 800-1200rpm, when it detects that the speed is lower than 850rpm for 10 consecutive minutes, combined with historical data analysis, a linear regression algorithm is used to predict the speed decline trend. If the predicted value is lower than 800 rpm in the next 5 minutes, the monitoring period is automatically adjusted from once every 30 seconds to once every 10 seconds, and the lower limit of the speed is temporarily raised to 850rpm to avoid the risk of too low speed.

[0161] Subsequently, the monitoring cycle of the pipeline pressure peak was optimized. The system set the pressure peak threshold to 1.6 MPa and used Fourier transform to analyze the frequency components of the pressure signal. If it was found that the high-frequency component accounted for more than 30%, it indicated that the pressure fluctuated violently. The monitoring cycle was automatically adjusted from once per minute to once every 15 seconds to ensure that the instantaneous peak was captured.

[0162] Finally, the system integrates these adjustment parameters to generate an adaptive safety monitoring strategy. Using machine learning models like support vector machines, it trains historical risk data and real-time data to output the optimal combination of monitoring frequency and speed range. For example, in high-risk scenarios, the recommended speed range is 900-1100 rpm, with a monitoring frequency of once every 10 seconds. This strategy is then pushed to the control center database in real time, creating a closed-loop monitoring system. To enhance logic, the system also integrates temperature monitoring. If the pipeline temperature exceeds 80 degrees Celsius, a dual pressure and speed check is automatically triggered, ensuring multi-dimensional safety control and establishing a rigorous safety monitoring network.

[0163] In summary, the present invention discloses a safety monitoring method for a ground-source heat pump system, which collects multi-dimensional real-time data such as the circulation pump speed, pipeline pressure, and ambient temperature through multi-point sensors, and realizes dynamic monitoring of the operating status through noise filtering and time series analysis. When an abnormality is detected, a comprehensive operation matrix is ​​constructed in combination with historical operation records, and the stress distribution of the underground pipeline is reversely calculated through structured grid division technology. Based on the stress distribution sequence, the present invention uses linear regression analysis to predict future stress evolution trends and determine the potential safety risk level. Finally, the data acquisition frequency is dynamically adjusted according to the risk level to generate an adaptive safety monitoring strategy. The present invention realizes intelligent safety monitoring of the ground-source heat pump system, can timely discover potential risks and take preventive measures, and effectively improve the operational safety and reliability of the system.

[0164] Reference Figure 2 , an embodiment of the present invention provides a structural diagram of a safety monitoring module of a ground source heat pump system, including:

[0165] The first acquisition module 201 is used to acquire a multi-dimensional data stream of the ground source heat pump system in real time, and obtain a real-time parameter set after filtering the multi-dimensional data stream according to a preset noise threshold;

[0166] A second acquisition module 202 is configured to determine a dynamic change trend based on the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, compare the real-time parameter set with associated fields in historical operation logs to obtain an abnormal time window;

[0167] A construction module 203 is configured to extract corresponding historical operation records according to the abnormal time window and generate a complete historical multidimensional data set, construct a comprehensive operation matrix according to the complete historical multidimensional data set, and obtain a stress distribution sequence according to the comprehensive operation matrix;

[0168] The generation module 204 is used to predict the stress evolution trend within a preset time period if the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, determine the potential safety risk signal level, dynamically adjust the acquisition frequency according to the potential safety risk signal level, and generate an adaptive safety monitoring strategy.

[0169] The embodiment of the present invention further provides a terminal device. The terminal device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned embodiments of the safety monitoring method for the ground source heat pump system are implemented, such as Figure 1 Alternatively, the processor implements the functions of the modules / units in the above-mentioned system embodiments when executing the computer program.

[0170] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0171] The terminal device may be a computing device such as a desktop computer, laptop, PDA, or smart tablet. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of terminal devices and do not constitute a limitation of the terminal device. The terminal device may include more or fewer components than those described above, or a combination of certain components or different components. For example, the terminal device may also include input / output devices, network access devices, buses, and the like.

[0172] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0173] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0174] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above-mentioned method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0175] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0176] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A safety monitoring method for a ground source heat pump system, characterized in that: include: Acquire a multidimensional data stream of the ground source heat pump system in real time, and filter the multidimensional data stream according to a preset noise threshold to obtain a real-time parameter set; Determining a dynamic change trend based on the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, comparing the real-time parameter set with associated fields of a historical operation log to obtain an abnormal time window; Extracting corresponding historical operation records according to the abnormal time window and generating a complete historical multidimensional data set, constructing a comprehensive operation matrix according to the complete historical multidimensional data set, and obtaining a stress distribution sequence according to the comprehensive operation matrix; If the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, the stress evolution trend within a preset time period is predicted, the potential safety risk signal level is determined, the acquisition frequency is dynamically adjusted according to the potential safety risk signal level, and an adaptive safety monitoring strategy is generated.

2. The safety monitoring method for a ground source heat pump system according to claim 1, characterized in that: The real-time acquisition of a multi-dimensional data stream of the ground source heat pump system and filtering the multi-dimensional data stream according to a preset noise threshold to obtain a real-time parameter set include: Acquire real-time data streams through multi-point sensors, remove abnormal values ​​exceeding the noise threshold, and obtain a preliminary data set; Smoothing the preliminary data set, and if it is detected that the ambient temperature fluctuation within a continuous time period exceeds a preset temperature fluctuation threshold, obtaining a corresponding abnormal data segment; Analyzing the trend of the abnormal data segment, if the correlation between the circulating pump speed and the pipeline pressure peak deviates from a normal range, adjusting the abnormal data to obtain an adjusted parameter set; The noise threshold is reset, the adjusted parameter set is calibrated, and it is determined whether it meets the requirements of the system operation state. If so, the real-time parameter set is obtained.

3. The safety monitoring method for a ground source heat pump system according to claim 1, characterized in that: Determining the dynamic change trend according to the real-time parameter set includes: Acquiring real-time data of the circulating pump from the real-time parameter set and performing smoothing processing to obtain smoothed speed sequence data and extracting characteristic values ​​of the fluctuation period therefrom; Comparing the characteristic value with a preset characteristic threshold, and if it exceeds, obtaining the abnormal fluctuation period and abnormal segment; Analyze the correlation between the abnormal fluctuation period and the dynamic change trend, dynamically evaluate the operating status of the circulation pump according to the correlation, and judge whether it meets the preset standard. If so, determine the dynamic change trend.

4. The safety monitoring method for a ground source heat pump system according to claim 3, characterized in that: If the dynamic change trend deviates from a preset normal range, comparing the real-time parameter set with the associated fields of the historical operation log to obtain an abnormal time window includes: If the dynamic change trend deviates from the preset normal range, obtaining the abnormal segment; Obtaining corresponding fields in the historical operation log according to the abnormal fragment and performing comparison to determine the abnormal time window; It then includes: extracting the fluctuation characteristics of the dynamic change trend according to the abnormal time window, and if the fluctuation characteristics continue to deviate from the preset fluctuation threshold, analyzing the operating status data in the time window to determine the specific distribution interval of the abnormality.

5. The safety monitoring method for a ground source heat pump system according to claim 1, characterized in that: The extracting corresponding historical operation records according to the abnormal time window and generating a complete historical multidimensional data set includes: Obtaining historical operation records of the abnormal time window to obtain an initial data set; If it is found that there is data missing in the initial data set, the historical data is used to complete the data and obtain the complete historical multidimensional data set.

6. The safety monitoring method for a ground source heat pump system according to claim 5, characterized in that: The comprehensive operation matrix is ​​constructed based on the complete historical multidimensional data set, including: Performing precision calibration on the complete historical multidimensional data set to obtain calibrated data and extracting influencing factors of ambient temperature fluctuations to generate a fluctuation feature set including the influencing factors; A comprehensive operation matrix including time series characteristics is generated according to the fluctuation characteristic set.

7. The safety monitoring method for a ground source heat pump system according to claim 6, characterized in that: Obtaining a stress distribution sequence according to the comprehensive operation matrix includes: Acquiring historical operation data of underground pipelines, and performing structured hierarchical organization on the historical operation data to obtain a hierarchical data set; Integrate the spatial mapping relationship of the historical operation data according to the hierarchical data set to generate an operation matrix; extracting an eigenvalue set related to stress distribution from the operation matrix, and correcting the eigenvalue set to obtain stress distribution data; If the stress distribution data deviates from a preset distribution threshold, the stress distribution data is grouped to generate the stress distribution sequence.

8. The safety monitoring method for a ground source heat pump system according to any one of claims 1 to 7, characterized in that: If the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, the stress evolution trend within a preset time period is predicted to determine the potential safety risk signal level, including: Obtain historical stress records of underground pipelines and extract stress change sets; Obtaining a stress change rate according to the stress change set, obtaining a stress change map according to the stress change rate and the historical stress record, and determining a prediction coefficient combination; If the coefficients in the prediction coefficient combination exceed a preset coefficient threshold, a stress evolution sequence is generated according to the prediction time span to obtain an evolution trend set; The evolution trend set is compared with the safety threshold to generate a safety risk sequence and determine the potential safety risk signal level.

9. The safety monitoring method for a ground source heat pump system according to any one of claims 1 to 7, characterized in that: The dynamically adjusting the acquisition frequency according to the potential safety risk signal level to generate an adaptive safety monitoring strategy includes: Obtaining an acquisition frequency reference value according to the potential safety risk signal level, dynamically adjusting the acquisition frequency according to the acquisition frequency reference value, generating a frequency adjustment plan, and determining an adjusted frequency range; If the adjusted frequency range exceeds the preset frequency threshold, a speed optimization sequence is generated in combination with the historical records of the circulating pump to obtain an optimized speed range; Dynamically matching the monitoring period of the pipeline pressure peak with the optimized speed range to generate a monitoring period plan, and determining whether the monitoring period plan meets the safety monitoring requirements; If so, the adjusted frequency range, the optimized speed range and the monitoring period scheme are integrated to generate the adaptive safety monitoring strategy.

10. A safety monitoring device for a ground source heat pump system, characterized in that: include: A first acquisition module is used to acquire a multi-dimensional data stream of the ground source heat pump system in real time, and obtain a real-time parameter set after filtering the multi-dimensional data stream according to a preset noise threshold; A second acquisition module is configured to determine a dynamic change trend based on the real-time parameter set, and if the dynamic change trend deviates from a preset normal range, compare the real-time parameter set with associated fields in a historical operation log to obtain an abnormal time window; a construction module, configured to extract corresponding historical operation records according to the abnormal time window and generate a complete historical multidimensional data set, construct a comprehensive operation matrix according to the complete historical multidimensional data set, and obtain a stress distribution sequence according to the comprehensive operation matrix; A generation module is used to predict the stress evolution trend within a preset time period if the stress distribution peak point in the stress distribution sequence exceeds the safety threshold, determine the potential safety risk signal level, dynamically adjust the acquisition frequency according to the potential safety risk signal level, and generate an adaptive safety monitoring strategy.

Images