A hydraulic lubrication system abnormality warning method and system for leak detection
By collecting and analyzing liquid level data in real time, using abnormal detection algorithms and multi-level early warning mechanisms to identify and deal with leakage in hydraulic lubrication systems, the problem of difficulty in detecting leakage in the liquid level monitoring system in the existing technology is solved, and the monitoring accuracy and Security.
Patent Information
- Application Number
- CN202411241828.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-05
AI Technical Summary
The existing liquid level monitoring system is difficult to detect leakage from hydraulic and lubricating systems in hot-rolled production lines in a timely manner, especially trace leakage or slow liquid level drop, and lacks efficient early warning mechanisms and emergency response strategies, resulting in increased production losses and safety risks. .
The liquid level sensor is used to collect data in real time, calculate the liquid level difference and set the initial compensation value, and use the abnormal detection algorithm to identify abnormal liquid level changes, adjust the compensation value based on historical data and seasonal fluctuations, set a multi-level early warning mechanism and adopt corresponding Emergency measures.
It improves the accuracy and reliability of liquid level monitoring, dynamically adapts to environmental changes, reduces false alarms and missed reports, promptly identify leakage signs, optimizes early warning mechanisms, reduces losses, and improves the safety of facilities and personnel.
Smart Images

Figure CN118757482B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of abnormality monitoring, and in particular to an abnormality early warning method and system for a hydraulic lubrication system for leakage detection. Background Art
[0002] In the operation of hot rolling production lines, the stable operation of hydraulic and lubrication systems is crucial, but due to the complexity and widespread distribution of pipelines, oil leakage incidents occur frequently and are difficult to detect in time. Traditionally, production lines rely on manual meter reading or high and low level alarm systems based on switch quantities to monitor liquid level changes. These methods have significant limitations, especially in terms of timeliness and accuracy. Once a leak occurs, it is often necessary to wait until the system is passively shut down before locating the leak point through manual investigation, which is not only time-consuming and labor-intensive, but also leads to significant production losses and economic damage.
[0003] Although manual meter reading is simple and easy, it relies on regular inspections and is easily affected by human factors, such as the accuracy and completeness of records. In addition, manual meter reading cannot achieve real-time monitoring, and it is often difficult to detect leaks that occur quickly, thus missing the best response time window.
[0004] Although the switch-type high and low level alarm system is an improvement over manual meter reading and can automatically sound an alarm when the liquid level reaches the preset high and low thresholds, this method has a fixed response threshold and lacks flexibility. It is difficult to respond effectively to small changes in the early stages of a leak. It is usually suitable for scenarios with more obvious changes in the liquid level. However, its monitoring capabilities are greatly reduced for trace leaks or slow drops in liquid levels.
[0005] Traditional liquid level monitoring systems may have inaccurate monitoring data due to factors such as sensor errors, inaccurate data processing or environmental changes, which will increase the risk of false alarms and missed alarms and affect reliability. At the same time, they lack the ability to dynamically adapt to environmental changes and different monitoring conditions, making it impossible to automatically adjust to maintain monitoring accuracy and stability when facing environmental factors such as temperature and pressure changes. When potential anomalies or leakage incidents are monitored, there is often a lack of effective early warning mechanisms and rapid response measures, resulting in the inability to take necessary countermeasures in a timely manner, increasing potential risks and losses. Due to the lack of efficient early warning mechanisms and emergency response strategies, existing liquid level monitoring systems perform poorly in managing potential safety risks and preventing accidents, affecting the safety of facilities and personnel.
[0006] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0007] In order to overcome the above problems, the present invention aims to propose a hydraulic lubrication system abnormal warning method and system for leakage detection, with the aim of solving the problem that due to the lack of efficient warning mechanism and emergency response strategy, the existing liquid level monitoring system performs poorly in managing potential safety risks and preventing accidents, thus affecting the safety of facilities and personnel.
[0008] To this end, the specific technical solution adopted by the present invention is as follows:
[0009] According to one aspect of the present invention, a hydraulic lubrication system abnormality early warning method for leakage detection is provided, and the hydraulic lubrication system abnormality early warning method comprises the following steps:
[0010] S1. Use the liquid level sensor to collect the liquid level data of the fuel tank in real time;
[0011] S2. Analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference;
[0012] S3. According to the abnormality detection algorithm, the original liquid level difference is adjusted using the initial liquid level difference compensation value to identify abnormal liquid level changes that deviate from the normal range;
[0013] S4. Adjust the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level changes and seasonal fluctuations of historical liquid level data;
[0014] S5. Evaluate the adjusted liquid level difference compensation value and abnormal detection results to determine whether there is leakage;
[0015] S6. Set up a multi-level early warning mechanism. When it is determined that there are signs of leakage, trigger the corresponding level of alarm according to the identification results and take corresponding emergency measures.
[0016] Optionally, analyzing the collected liquid level data, calculating the liquid level difference within a preset time period, and setting an initial liquid level difference compensation value according to the liquid level difference comprises the following steps:
[0017] S21, collecting liquid level data within a preset time period;
[0018] S22, calculating the liquid level difference within the preset time period by comparing the liquid level readings at the beginning and end of the time period according to the liquid level data within the preset time period;
[0019] S23, analyzing the trend of liquid level change according to the calculated liquid level difference, and determining the normal liquid level change range;
[0020] S24, setting an initial liquid level difference compensation value according to the analysis results and the normal liquid level variation range;
[0021] S25. Record the calculated liquid level difference and the set initial liquid level difference compensation value.
[0022] Optionally, according to the abnormality detection algorithm, the original liquid level difference is adjusted using the initial liquid level difference compensation value, and the identification of abnormal liquid level changes that deviate from the normal range includes the following steps:
[0023] S31, obtaining the real-time collected liquid level data, and combining the adjusted liquid level difference with the factors affecting the liquid level change to construct a multivariate time series data set;
[0024] S32, constructing a neighborhood graph based on the data at each time point in the multivariate time series data set;
[0025] S33, using a Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph to calculate the weight, and construct a weight matrix;
[0026] S34, calculating the transfer matrix and the transfer vector according to the weight matrix and the degree matrix of the liquid level data at the time point;
[0027] S35, using the ADPP algorithm formula to calculate the connectivity of each data point;
[0028] S36: Based on the calculated connectivity, and comparing it with a preset threshold, identifying a time point when the connectivity is lower than the preset threshold.
[0029] Optionally, calculating the weights by using a Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph and constructing a weight matrix includes the following steps:
[0030] S331, define a Gaussian function and calculate the weight between two points;
[0031] S332, for each pair of liquid level data at a time point in the neighborhood graph, calculating the Euclidean distance between each pair of liquid level data at a time point;
[0032] S333, using a Gaussian function to calculate the weight between each pair of liquid level data at each time point according to the Euclidean distance;
[0033] S334, filling all calculated weights into the weight matrix to construct the weight matrix;
[0034] Among them, the weight formula between each pair of liquid level data at time points is:
[0035] ;
[0036] In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between
[0037] represents the natural exponential function;
[0038] Indicates the liquid level data at a point in time i Liquid level data at time points j The Euclidean distance between
[0039] represents the standard deviation of the Gaussian function;
[0040] Represents the part of the Gaussian function that controls the speed of weight decay.
[0041] Optionally, calculating the connectivity of each data point using the ADPP algorithm formula includes the following steps:
[0042] S351, for each time point of the liquid level data in the multivariate time series data set, determine the liquid level data of all neighboring time points of the liquid level data at each time point in the neighborhood graph;
[0043] S352, for the liquid level data at each time point, calculating the connectivity of the liquid level data at each time point;
[0044] S353, recording the connectivity of the liquid level data at each time point, and comparing it with a preset threshold;
[0045] Among them, the formula for calculating the connectivity of the liquid level data at each time point is:
[0046] ;
[0047] In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between
[0048] k Represents the liquid level data at adjacent time points;
[0049] Indicates the liquid level data at a point in time i Connectivity
[0050] i , j All represent the liquid level data at a time point.
[0051] Optionally, adjusting the initial liquid level difference compensation value according to the identified abnormal liquid level change and the analysis result of the seasonal fluctuation of the historical liquid level data includes the following steps:
[0052] S41, evaluating the change, amplitude and duration of the abnormal liquid level according to the time point when the connectivity is lower than the preset threshold;
[0053] S42, comparing the abnormal liquid level change with the historical data, combining the seasonal fluctuations of the historical liquid level data, identifying whether there are common abnormalities under specific conditions, and determining whether the common abnormalities are in the initial liquid level difference compensation value;
[0054] S43, if it is found that the abnormal liquid level changes are concentrated under specific conditions, the initial liquid level difference compensation value is adjusted under the specific conditions;
[0055] S44, if the abnormal liquid level change is globally distributed, globally adjust the initial liquid level difference compensation value;
[0056] S45. Re-perform abnormality detection using the adjusted liquid level difference compensation value, and record each adjusted liquid level difference compensation value and the corresponding abnormality detection result.
[0057] Optionally, evaluating the adjusted level difference compensation value and the abnormal detection result to determine whether there is a leakage sign includes the following steps:
[0058] S51, analyzing whether the abnormal detection result matches the known leakage event according to the liquid level difference compensation value after each adjustment and the corresponding abnormal detection result;
[0059] S52, comparing the currently detected anomaly with the historical anomaly data in detail to evaluate whether there are repeated anomaly patterns and whether the anomaly patterns have been associated with leakage events in the past;
[0060] S53, by comparing the abnormal detection results before and after the adjustment, determining whether false positives or false negatives are reduced;
[0061] S54: Based on the evaluation results of steps S51 to S53, if the detected anomaly conforms to a known leakage pattern and is consistent with historical leakage data, it is determined that there is a sign of leakage.
[0062] Optionally, a multi-level early warning mechanism is set. When it is determined that there are signs of leakage, an alarm of a corresponding level is triggered according to the identification result, and corresponding emergency measures are taken, including the following steps:
[0063] S61. Based on the severity of the leakage signs detected in real time, define multiple warning levels and formulate emergency response measures for each warning level, wherein the multiple warning levels include a low warning level, a medium warning level and a high warning level;
[0064] S62, using the obtained real-time anomaly detection results, and classifying the anomaly detection results into corresponding warning levels according to severity;
[0065] S63, matching the corresponding leakage pattern according to the type and characteristics of the real-time anomaly;
[0066] S64. If the abnormal detection result is identified and classified as a specific warning level, the warning signal corresponding to the corresponding level and the pre-set emergency response measures are automatically triggered;
[0067] S65. Implement emergency response measures that match the triggered warning level.
[0068] Optionally, if the abnormal detection result is identified and classified as a specific warning level, automatically triggering a warning signal corresponding to the corresponding level and a pre-set emergency response measure includes the following steps:
[0069] S641. Configure a unique alarm signal for each warning level;
[0070] S642. Based on the leakage pattern matching result, define a triggered warning level for each type and characteristic of anomaly;
[0071] S643. Establish an automated response mechanism to transmit warning signals and execute emergency response measures when warnings are triggered;
[0072] S644. Collect feedback information when implementing emergency response measures, and use the feedback information to evaluate the effectiveness of the response measures and identify problems in implementation.
[0073] According to another aspect of the present invention, there is also provided a hydraulic lubrication system abnormality warning system for leakage detection, the system comprising: a liquid level data acquisition module, a liquid level data analysis module, an abnormality detection adjustment module, a liquid level difference compensation adjustment module, a leakage sign evaluation module and a multi-level warning mechanism module;
[0074] The liquid level data acquisition module is connected to the abnormality detection adjustment module through the liquid level data analysis module, the abnormality detection adjustment module is connected to the leakage sign evaluation module through the liquid level difference compensation adjustment module, and the leakage sign evaluation module is connected to the multi-level early warning mechanism module;
[0075] Liquid level data acquisition module, used to collect the liquid level data of the oil tank in real time using the liquid level sensor;
[0076] The liquid level data analysis module is used to analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference;
[0077] An abnormality detection adjustment module is used to adjust the original liquid level difference using the initial liquid level difference compensation value according to the abnormality detection algorithm, and identify abnormal liquid level changes that deviate from the normal range;
[0078] A liquid level difference compensation adjustment module is used to adjust the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level changes and seasonal fluctuations of historical liquid level data;
[0079] A leakage sign evaluation module is used to evaluate the adjusted liquid level difference compensation value and abnormal detection results to determine whether there is leakage sign;
[0080] The multi-level early warning mechanism module is used to set up a multi-level early warning mechanism. When it is determined that there are signs of leakage, the corresponding level of alarm is triggered according to the identification results, and corresponding emergency measures are taken.
[0081] Compared with the prior art, this application has the following beneficial effects:
[0082] 1. The present invention can significantly improve the accuracy of liquid level monitoring by accurately calculating the liquid level difference and setting a reasonable initial compensation value, reduce the impact of factors such as sensor errors and environmental changes, analyze the liquid level change trend and adjust the compensation value accordingly, and can dynamically adapt to different monitoring conditions and environmental changes to improve robustness. By setting the initial compensation value, potential abnormal liquid level changes can be more accurately identified, thereby optimizing the early warning mechanism, taking necessary countermeasures in advance, and reducing possible losses. Recording and analyzing the liquid level difference and compensation value provides a continuous optimization monitoring strategy, supports data-based decision-making, and improves the accuracy and efficiency of decision-making.
[0083] 2. Through comprehensive analysis of multivariate time series data, the algorithm of the present invention can more accurately identify abnormal liquid level changes that deviate from the normal range. The initial liquid level difference compensation value is used for adjustment so that the monitoring can dynamically adapt to environmental changes and data fluctuations, thereby improving adaptability and flexibility. The constructed multivariate time series data set and the analysis through the ADPP algorithm provide data-driven support for decision-making, which is helpful for accurately identifying and quickly responding to abnormal situations. By accurately calculating the connectivity of data points and comparing with preset thresholds, the possibility of false alarms and missed alarms can be effectively reduced, and reliability can be enhanced. Once an abnormality is identified, timely measures can be taken to deal with it, thereby reducing potential risks and losses.
[0084] 3. The present invention provides a basis for in-depth understanding of abnormal events by evaluating in detail the time points when the connectivity is lower than the preset threshold and analyzing the changes, amplitude and duration of abnormal liquid levels. Comparing the abnormal liquid level changes with historical data, especially considering seasonal fluctuations, helps to identify common anomalies under specific conditions, thereby providing a basis for adjusting the compensation value. The compensation value is finely adjusted according to the specific conditions or global distribution of the abnormal concentration, which helps to reduce false alarms and improve the accuracy of anomaly detection. Re-detecting anomalies using the adjusted compensation value is an important step in verifying the effectiveness of the compensation value adjustment, thereby ensuring the correctness and effectiveness of the adjustment measures.
[0085] 4. The present invention can accurately determine whether the current anomaly conforms to the leakage pattern by comparing the adjusted anomaly detection results with known leakage events. Comparing the current anomaly with historical anomaly data is helpful to identify repeated anomaly patterns and improve the accuracy and timeliness of leakage detection. By comparing the detection results before and after adjustment, the impact of the compensation value adjustment on reducing false alarms or missed alarms can be intuitively seen, thereby optimizing the overall performance. Through accurate anomaly detection and leakage sign judgment, potential leakage events can be identified in a timely manner, providing a basis for taking rapid response measures.
[0086] 5. The multi-level early warning mechanism and automated emergency response of the present invention ensure that actions can be taken quickly and effectively when signs of leakage are identified, thereby minimizing or preventing potential losses and injuries. Through timely early warning and emergency measures, the safety of facilities and personnel can be significantly improved, possible accidents and health risks can be avoided, and the early warning level classification and emergency response execution based on real-time abnormal detection results enhance data support for decision-making, ensure the scientificity and rationality of measures, and continuously learn and optimize by collecting feedback information and evaluating the effectiveness of emergency response, so as to improve the efficiency and effectiveness of future event response. BRIEF DESCRIPTION OF THE DRAWINGS
[0087] The above characteristics, features and advantages of the present invention and their implementation methods and methods will become more clearly understood in conjunction with the following description of the embodiments, which are described in detail in conjunction with the accompanying drawings. Herein, a schematic diagram is shown:
[0088] Figure 1 is a flow chart of a hydraulic lubrication system abnormality warning method for leakage detection according to an embodiment of the present invention;
[0089] Figure 2 The invention is a principle block diagram of a hydraulic lubrication system abnormality warning system for leakage detection according to an embodiment of the present invention.
[0090] In the figure:
[0091] 1. Liquid level data acquisition module; 2. Liquid level data analysis module; 3. Abnormal detection and adjustment module; 4. Liquid level difference compensation and adjustment module; 5. Leakage sign assessment module; 6. Multi-level early warning mechanism module. DETAILED DESCRIPTION
[0092] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0093] According to an embodiment of the present invention, a hydraulic lubrication system abnormality warning method and system for leakage detection are provided.
[0094] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a hydraulic lubrication system abnormality warning method for leakage detection is provided, and the hydraulic lubrication system abnormality warning method comprises the following steps:
[0095] S1. Use the liquid level sensor to collect the liquid level data of the fuel tank in real time.
[0096] It needs to be explained that, first of all, it is necessary to select a suitable liquid level sensor according to the characteristics of the fuel tank (such as capacity, type of liquid used, environmental conditions, etc.); commonly used types of liquid level sensors include ultrasonic sensors, float sensors and capacitive sensors, etc. The sensor should be installed in a position that can accurately measure the liquid level in the tank. The installation process needs to ensure that the sensor is stable and sensitive to changes in the liquid level, and the collected data is transmitted in real time. The liquid level sensor begins to monitor the liquid level in the fuel tank in real time and sends the data to the database. The data will be recorded and used for subsequent analysis. The collected data needs to be transmitted to the database for storage.
[0097] The choice of sensor directly affects the accuracy of the data and the reliability of the system. For example, for volatile liquids, it may be more appropriate to use a non-contact ultrasonic sensor to avoid safety hazards; while for opaque liquids or in the presence of foam, choosing a capacitive sensor provides more stable performance.
[0098] S2. Analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference.
[0099] Preferably, analyzing the collected liquid level data, calculating the liquid level difference within a preset time period, and setting an initial liquid level difference compensation value according to the liquid level difference comprises the following steps:
[0100] S21, collecting liquid level data within a preset time period, first using the installed liquid level sensor to continuously monitor and record the liquid level data of the oil tank within a preset time period (such as every hour, every day);
[0101] S22, calculating the liquid level difference within the preset time period by comparing the liquid level readings at the beginning and end of the time period according to the liquid level data within the preset time period;
[0102] By comparing the liquid level readings at the beginning and end of the time period, and then calculating the liquid level change during this period, that is, the liquid level difference, extract the starting and ending liquid level readings of the time period, and finally calculate the difference between the two.
[0103] S23, analyzing the trend of liquid level change according to the calculated liquid level difference, and determining the normal liquid level change range;
[0104] Use statistical analysis methods (such as mean, standard deviation) to analyze the calculated liquid level difference and combine it with historical data to determine the normal range of liquid level changes.
[0105] S24, setting an initial liquid level difference compensation value according to the analysis results and the normal liquid level variation range;
[0106] Based on the liquid level change trend and normal change range, the initial liquid level difference compensation value is set, and the setting of the compensation value needs to comprehensively consider multiple factors, including system errors, environmental factors, etc.
[0107] S25. Record the calculated liquid level difference and the set initial liquid level difference compensation value.
[0108] It should be explained that historical data is used to analyze the common patterns and ranges of liquid level changes, and the errors that may be introduced during data collection and processing, such as sensor accuracy, data transmission delays, etc., are analyzed. The impact of environmental factors on liquid level changes, such as temperature and pressure changes, is considered. Based on the above factors, a reasonable initial compensation value is set to optimize the accuracy of liquid level change judgment. The compensation value is actually applied and the results are observed to verify its effect on improving detection accuracy. The calculated liquid level difference each time and its corresponding initial liquid level difference compensation value are recorded in the database for subsequent analysis and adjustment.
[0109] S3. According to the abnormality detection algorithm, the original liquid level difference is adjusted using the initial liquid level difference compensation value to identify abnormal liquid level changes that deviate from the normal range.
[0110] Preferably, according to the abnormality detection algorithm, the original liquid level difference is adjusted using the initial liquid level difference compensation value, and the identification of abnormal liquid level changes that deviate from the normal range includes the following steps:
[0111] S31. Acquire the real-time collected liquid level data, and combine the adjusted liquid level difference with the factors affecting the liquid level change to construct a multivariate time series data set.
[0112] Use database management for data storage and use data processing software (such as Python pandas library) for data preprocessing and organization. First, obtain the real-time liquid level data. Then, combine this data with other factors that affect the liquid level change (such as temperature, pressure, historical leakage events, etc.) to build a multivariate time series data set. The multivariate time series data set should include the timestamp of each observation point and the values of all related variables.
[0113] S32. Construct a neighborhood graph based on the data at each time point in the multivariate time series data set.
[0114] Use a machine learning library (such as Python's Scikit-learn or similar libraries) to calculate the distance between data points and build a neighborhood graph. Common distance metrics include Euclidean distance, Manhattan distance, etc. Use a multivariate time series dataset to build a neighborhood graph by calculating the similarity or distance between data points; a neighborhood graph is a graphical representation in which each data point is a node in the graph, and the similarity or distance between data points determines whether there are edges connecting the nodes.
[0115] S33. Use the Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph to calculate the weight and construct a weight matrix.
[0116] Preferably, calculating the weights by using a Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph and constructing a weight matrix comprises the following steps:
[0117] S331, define a Gaussian function and calculate the weight between two points;
[0118] S332, for each pair of liquid level data at a time point in the neighborhood graph, calculating the Euclidean distance between each pair of liquid level data at a time point;
[0119] S333, using a Gaussian function to calculate the weight between each pair of liquid level data at each time point according to the Euclidean distance;
[0120] First, the Euclidean distance between each pair of nodes in the neighborhood graph is calculated. For each pair of nodes, the distance between them is substituted into the Gaussian function to calculate the weight. The Gaussian function is a decay function that reduces the weight value as the distance increases, which helps to emphasize the relationship between nodes that are close to each other. The value is crucial to the effect of the Gaussian function. Too large or too small a value may affect the accuracy and effect of weight calculation.
[0121] S334, filling all calculated weights into the weight matrix to construct the weight matrix;
[0122] Among them, the weight formula between each pair of liquid level data at time points is:
[0123] ;
[0124] In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between
[0125] represents the natural exponential function;
[0126] Indicates the liquid level data at a point in time i Liquid level data at time points j The Euclidean distance between
[0127] represents the standard deviation of the Gaussian function;
[0128] Represents the part of the Gaussian function that controls the speed of weight decay.
[0129] In addition, it should be explained that the Gaussian function (also called the Gaussian kernel) is usually used to measure the similarity between points. First, the Gaussian function is defined to control the speed at which the weight changes with distance. For each pair of nodes in the neighborhood graph, the Euclidean distance between them is calculated. In order to prepare the data for the application of the Gaussian function, the Euclidean distance between vectors is calculated using a library such as numpy in Python. The Euclidean distance between each pair of liquid level data at each time point calculated in the previous step is substituted into the Gaussian function, and the weights between these point pairs are calculated. All the calculated weight values are filled into a matrix to form a weight matrix. Each element of the weight matrix represents the weight between a pair of data points.
[0130] S34. Calculate the transfer matrix and the transfer vector according to the weight matrix and the degree matrix of the liquid level data at the time point.
[0131] Using the weight matrix constructed previously, calculate the degree of each node, that is, the sum of the weights of all connections of the node; the degree matrix of a node is a diagonal matrix, and each element on its diagonal represents the degree of a node. The degree of the node is calculated by simply summing each row or column of the weight matrix and placing the result in the corresponding position of the diagonal matrix. The transfer matrix is calculated by the weight matrix and the degree matrix of the node. Each element in the transfer matrix is defined as the node i To Node j The transition probability can be obtained by dividing each element of the weight matrix by the node i The transfer vector can be calculated from the transfer matrix, which represents the transition probability from one node to all other nodes.
[0132] S35. Calculate the connectivity of each data point using the ADPP algorithm formula.
[0133] Preferably, calculating the connectivity of each data point using the ADPP algorithm formula includes the following steps:
[0134] S351, for each time point of the liquid level data in the multivariate time series data set, determine the liquid level data of all neighboring time points of the liquid level data at each time point in the neighborhood graph;
[0135] S352, for the liquid level data at each time point, calculating the connectivity of the liquid level data at each time point;
[0136] S353, recording the connectivity of the liquid level data at each time point, and comparing it with a preset threshold;
[0137] Among them, the formula for calculating the connectivity of the liquid level data at each time point is:
[0138] ;
[0139] In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between
[0140] k Represents the liquid level data at adjacent time points;
[0141] Indicates the liquid level data at a point in time i Connectivity
[0142] i , j All represent the liquid level data at a time point.
[0143] Determine all neighboring time points of the liquid level data at each time point in the neighborhood graph, and find all other data points connected to each data point. This can be achieved by analyzing the weight matrix. The position of the non-zero element in the weight matrix indicates the existence of connected data points. The connectivity of the liquid level data at each time point is calculated through the connectivity formula, which takes into account the weight between the data point and its neighbors. The connectivity of the liquid level data at each time point is recorded and compared with a preset threshold to identify data points whose connectivity is significantly lower than other points. Data points lower than other points are used to indicate abnormalities.
[0144] S36: Based on the calculated connectivity, and comparing it with a preset threshold, identifying a time point when the connectivity is lower than the preset threshold.
[0145] It should be explained that based on the calculated and recorded connectivity, the time points where the connectivity is below the preset threshold are identified. The time points represent abnormal or unusual liquid level changes, which may be signs of leakage. This is achieved by traversing all recorded connectivity data and comparing them with the preset threshold. The identified low connectivity time points can be further analyzed or trigger alarms.
[0146] The ADPP algorithm is a process used to process specific types of data (such as multivariate time series data sets) and perform specific tasks (such as anomaly detection) based on this. The ADPP algorithm is obviously used to analyze liquid level data, especially to identify abnormal liquid level changes that deviate from the normal fluctuation range; the core of the ADPP algorithm is to identify abnormal patterns in the data through data processing and analysis methods. It is suitable for fields such as liquid level monitoring, where the analysis of multivariate and time series data is crucial for early warning and problem diagnosis. By calculating the connectivity between data points and comparing them with preset thresholds, the algorithm can effectively identify abnormal liquid level changes, thereby providing a basis for taking appropriate countermeasures.
[0147] S4. Adjust the initial liquid level difference compensation value based on the identified abnormal liquid level changes and the analysis results of seasonal fluctuations in historical liquid level data.
[0148] Preferably, adjusting the initial liquid level difference compensation value according to the identified abnormal liquid level change and the analysis result of the seasonal fluctuation of the historical liquid level data comprises the following steps:
[0149] S41. Evaluate the change, amplitude and duration of the abnormal liquid level according to the time point at which the connectivity is lower than a preset threshold.
[0150] This is achieved by comparing the liquid level data of the abnormal point with the data of the time points before and after it. Statistical analysis methods (such as calculating the mean, median, standard deviation, etc. of the changes) are used to evaluate the liquid level change characteristics of the abnormal point in detail according to the time points when the identified connectivity is lower than the preset threshold, including the magnitude of the change, direction (liquid level rise or fall) and duration.
[0151] S42. Compare the abnormal liquid level changes with historical data, and combine the seasonal fluctuations of the historical liquid level data to identify whether there are common anomalies under specific conditions, and determine whether the common anomalies are in the initial liquid level difference compensation value.
[0152] The currently identified abnormal level changes are compared with historical level data, especially considering the impact of seasonal fluctuations, in order to identify whether these anomalies belong to common patterns under certain specific conditions.
[0153] S43. If it is found that the abnormal liquid level changes are concentrated under specific conditions, the initial liquid level difference compensation value is adjusted under the specific conditions.
[0154] If it is found that abnormal liquid level changes are mainly concentrated under specific conditions (such as a specific season or specific operating conditions), the initial liquid level difference compensation value is adjusted under these conditions, and then the compensation value is fine-tuned according to the characteristics of the abnormal liquid level changes to reduce false alarms under these conditions.
[0155] S44. If the abnormal liquid level change is globally distributed, the initial liquid level difference compensation value is globally adjusted, and a new global liquid level difference compensation value can be re-evaluated and set based on a comprehensive analysis of all identified abnormal points.
[0156] S45. Re-perform abnormality detection using the adjusted liquid level difference compensation value, and record each adjusted liquid level difference compensation value and the corresponding abnormality detection result.
[0157] It should be explained that the data is re-detected for anomalies using the adjusted level difference compensation value to verify the effectiveness of the adjustment, the previous anomaly detection process is repeated, and the new detection results are compared with the previous results to evaluate the impact of the compensation value adjustment.
[0158] In the above steps, the key is to be able to correctly distinguish the reasons behind the abnormal data, whether it is normal fluctuations caused by environmental conditions or potential problems. By comparing with historical data and analyzing seasonal fluctuations, the compensation value can be adjusted more accurately, thereby improving the accuracy of identifying abnormal conditions.
[0159] S5. Evaluate the adjusted liquid level difference compensation value and abnormal detection results to determine whether there is any sign of leakage.
[0160] Preferably, evaluating the adjusted liquid level difference compensation value and the abnormal detection result to determine whether there is leakage includes the following steps:
[0161] S51, analyzing whether the abnormal detection result matches the known leakage event according to the liquid level difference compensation value after each adjustment and the corresponding abnormal detection result;
[0162] S52, comparing the currently detected anomaly with the historical anomaly data in detail to evaluate whether there are repeated anomaly patterns and whether the anomaly patterns have been associated with leakage events in the past;
[0163] S53, by comparing the abnormal detection results before and after the adjustment, determining whether false positives or false negatives are reduced;
[0164] S54: Based on the evaluation results of steps S51 to S53, if the detected anomaly conforms to a known leakage pattern and is consistent with historical leakage data, it is determined that there is a sign of leakage.
[0165] It needs to be explained that the abnormal detection results corresponding to the liquid level difference compensation value after each adjustment are analyzed, and these results are compared with known leakage events to determine the accuracy and relevance of the abnormal detection. The currently detected abnormalities are carefully compared with the historical abnormal data to evaluate whether there are repeated abnormal patterns and analyze whether these patterns have been associated with leakage events in the past. The effectiveness of the adjustment measures can be evaluated by statistical analysis and comparison of detection results in different time periods. The abnormal detection results before and after the adjustment of the liquid level difference compensation value are compared to evaluate whether the adjustment helps to reduce false alarms or missed alarms. Based on the evaluation results of the above steps, if the detected anomaly conforms to a known leakage pattern and is consistent with historical leakage data, it can be determined that there are signs of leakage.
[0166] The key to the above steps is to be able to accurately identify recurring abnormal patterns and understand the relationship between these patterns and past leakage incidents. This requires a deep understanding of historical data and the ability to discover patterns and trends to improve the early warning capabilities of potential leakage incidents.
[0167] S6. Set up a multi-level early warning mechanism. When it is determined that there are signs of leakage, trigger the corresponding level of alarm according to the identification results and take corresponding emergency measures.
[0168] Preferably, a multi-level early warning mechanism is set up. When it is determined that there are signs of leakage, an alarm of a corresponding level is triggered according to the identification result, and corresponding emergency measures are taken, including the following steps:
[0169] S61. Based on the severity of leakage signs detected in real time, define multiple warning levels and formulate emergency response measures for each warning level, where the multiple warning levels include low warning level, medium warning level and high warning level.
[0170] S62. Utilize the obtained real-time anomaly detection results and classify the anomaly detection results into corresponding warning levels according to the severity.
[0171] S63. Match the corresponding leakage pattern according to the type and characteristics of the real-time anomaly.
[0172] S64. If the abnormal detection result is identified and classified as a specific warning level, the warning signal corresponding to the corresponding level and the pre-set emergency response measures are automatically triggered.
[0173] Preferably, if the abnormal detection result is identified and classified as a specific warning level, automatically triggering the warning signal corresponding to the corresponding level and the pre-set emergency response measures includes the following steps:
[0174] S641. Configure a unique alarm signal for each warning level;
[0175] Setting different alarm signals (such as different sounds, lights or message prompts) to distinguish low, medium and high warning levels helps ensure that personnel can quickly identify the warning level and take appropriate actions.
[0176] S642. Based on the leakage pattern matching result, define a triggered warning level for each type and characteristic of anomaly;
[0177] According to the specific characteristics of the anomaly detection results and their matching degree with the known leakage patterns, a corresponding warning level is defined for each anomaly.
[0178] S643. Establish an automated response mechanism to transmit warning signals and execute emergency response measures when warnings are triggered;
[0179] When an abnormality of a specific warning level is identified, the warning signal is automatically transmitted and pre-set emergency response measures are executed, including sending notifications, activating safety systems, or shutting down specific equipment.
[0180] S644. Collect feedback information when implementing emergency response measures, and use the feedback information to evaluate the effectiveness of the response measures and identify problems in implementation.
[0181] When implementing emergency response measures, feedback information is collected at the same time, such as response time, personnel reaction, equipment status, etc., to evaluate the effectiveness of the emergency response and identify possible implementation problems. Feedback information can be collected and evaluated through monitoring systems and subsequent analysis reports. Implementing a feedback mechanism involves the use of data recording tools and analysis software, as well as the establishment of a standard process for collecting feedback.
[0182] S65. Implement emergency response measures that match the triggered warning level.
[0183] It needs to be explained that, according to the severity of the signs of leakage, multiple warning levels (such as low, medium and high) are defined, and specific emergency response measures are formulated for each level, including analysis of possible leakage scenarios, potential impacts and emergency resources, and determining the warning level based on the parameters of the anomaly detection results (such as the magnitude and speed of change). According to the severity of the real-time anomaly detection results, they are classified into the corresponding warning level, and according to the type and characteristics of the real-time anomaly, the corresponding leakage pattern is matched to determine the appropriate response measures. When the anomaly detection result is identified and classified as a specific warning level, the warning signal corresponding to that level and the preset emergency response measures are automatically triggered. According to the triggered warning level, the matching emergency response measures are executed, such as evacuating personnel, shutting down related equipment or notifying the maintenance team.
[0184] According to another embodiment of the present invention, Figure 2As shown, a hydraulic lubrication system abnormality warning system for leakage detection is also provided, the system comprising: a liquid level data acquisition module 1, a liquid level data analysis module 2, an abnormality detection adjustment module 3, a liquid level difference compensation adjustment module 4, a leakage sign evaluation module 5 and a multi-level warning mechanism module 6;
[0185] The liquid level data acquisition module 1 is connected to the abnormality detection and adjustment module 3 through the liquid level data analysis module 2, the abnormality detection and adjustment module 3 is connected to the leakage sign evaluation module 5 through the liquid level difference compensation adjustment module 4, and the leakage sign evaluation module 5 is connected to the multi-level early warning mechanism module 6;
[0186] The liquid level data acquisition module 1 is used to collect the liquid level data of the oil tank in real time using the liquid level sensor;
[0187] Liquid level data analysis module 2, used to analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference;
[0188] The abnormality detection adjustment module 3 is used to adjust the original liquid level difference using the initial liquid level difference compensation value according to the abnormality detection algorithm, and identify abnormal liquid level changes that deviate from the normal range;
[0189] The liquid level difference compensation adjustment module 4 is used to adjust the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level changes and the seasonal fluctuations of the historical liquid level data;
[0190] A leakage sign evaluation module 5 is used to evaluate the adjusted liquid level difference compensation value and the abnormal detection result to determine whether there is a leakage sign;
[0191] The multi-level warning mechanism module 6 is used to set a multi-level warning mechanism. When it is determined that there are signs of leakage, an alarm of a corresponding level is triggered according to the identification result, and corresponding emergency measures are taken.
[0192] In summary, with the help of the above technical solutions of the present invention, the present invention can significantly improve the accuracy of liquid level monitoring by accurately calculating the liquid level difference and setting a reasonable initial compensation value, reduce the influence caused by factors such as sensor errors and environmental changes, analyze the liquid level change trend and adjust the compensation value accordingly, can dynamically adapt to different monitoring conditions and environmental changes, and improve robustness. By setting the initial compensation value, potential abnormal liquid level changes can be more accurately identified, thereby optimizing the early warning mechanism, taking necessary countermeasures in advance, and reducing possible losses. Recording and analyzing the liquid level difference and compensation value provides a continuous optimization monitoring strategy, supports data-based decision-making, and improves the accuracy and efficiency of decision-making; the present invention comprehensively analyzes multivariate time series data, and the algorithm can more accurately identify abnormal liquid level changes that deviate from the normal range. The initial liquid level difference compensation value is used for adjustment so that the monitoring can dynamically adapt to environmental changes and data fluctuations, improve adaptability and flexibility, and the constructed multivariate time series data set and the analysis through the ADPP algorithm provide data-driven support for decision-making, which is helpful to accurately identify and quickly respond to abnormal situations. By accurately calculating the connectivity of data points and comparing preset thresholds, the possibility of false alarms and missed alarms can be effectively reduced, and reliability is enhanced. Once an abnormality is identified, Measures can be taken in time to deal with it, thereby reducing potential risks and losses; the present invention provides a basis for in-depth understanding of abnormal events by evaluating in detail the time point when the connectivity is lower than the preset threshold, analyzing the change, amplitude and duration of the abnormal liquid level, and comparing the abnormal liquid level change with historical data, especially considering seasonal fluctuations, which helps to identify common anomalies under specific conditions, thereby providing a basis for adjusting the compensation value. According to the specific conditions or global distribution of the abnormal concentration, the compensation value is finely adjusted, which helps to reduce false alarms and improve the accuracy of abnormal detection. The adjusted compensation value is used to re-detect abnormalities, which is an important link to verify the effectiveness of the compensation value adjustment, ensuring the correctness and effectiveness of the adjustment measures; by comparing the adjusted abnormal detection results with known leakage events, the present invention can accurately determine whether the current anomaly conforms to the leakage pattern, and compare the current anomaly with historical abnormal data, which helps to identify repeated abnormal patterns and improve the accuracy and timeliness of leakage detection. By comparing the detection results before and after the adjustment, the impact of the compensation value adjustment on reducing false alarms or missed alarms can be intuitively seen, optimizing the overall performance, and through accurate abnormal detection and leakage sign judgment, potential leakage events can be identified in time, providing a basis for taking rapid response measures;The multi-level early warning mechanism and automated emergency response of the present invention ensure that actions can be taken quickly and effectively when signs of leakage are identified, thereby minimizing or preventing potential losses and injuries. Through timely early warning and emergency measures, the safety of facilities and personnel can be significantly improved, and possible accidents and health risks can be avoided. The early warning level classification and emergency response execution based on real-time abnormal detection results enhance the data support for decision-making, ensure the scientificity and rationality of measures, and continuously learn and optimize by collecting feedback information and evaluating the emergency response effect, so as to improve the efficiency and effectiveness of future incident response. ;
[0193] Although the present invention has been disclosed as above with preferred embodiments, the embodiments are merely examples for the convenience of description and are not intended to limit the present invention. Those skilled in the art may make several changes and modifications without departing from the spirit and scope of the present invention. The scope of protection claimed by the present invention shall be based on the claims.
Claims
1. A hydraulic lubrication system abnormality warning method for leakage detection, characterized in that: The abnormality warning method of the hydraulic lubrication system comprises the following steps: S1. Use the liquid level sensor to collect the liquid level data of the fuel tank in real time; S2. Analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference; S3. According to the abnormality detection algorithm, the original liquid level difference is adjusted using the initial liquid level difference compensation value to identify abnormal liquid level changes that deviate from the normal range; S4. Adjust the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level changes and seasonal fluctuations of historical liquid level data; S5. Evaluate the adjusted liquid level difference compensation value and abnormal detection results to determine whether there is leakage; S6. Set up a multi-level early warning mechanism. When it is determined that there are signs of leakage, trigger the corresponding level of alarm according to the identification results and take corresponding emergency measures; The method of adjusting the original liquid level difference using the initial liquid level difference compensation value according to the abnormality detection algorithm and identifying abnormal liquid level changes that deviate from the normal range includes the following steps: S31, obtaining the real-time collected liquid level data, and combining the adjusted liquid level difference with the factors affecting the liquid level change to construct a multivariate time series data set; S32, constructing a neighborhood graph based on the data at each time point in the multivariate time series data set; S33, using a Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph to calculate the weight, and construct a weight matrix; S34, calculating the transfer matrix and the transfer vector according to the weight matrix and the degree matrix of the liquid level data at the time point; S35, using the ADPP algorithm formula to calculate the connectivity of each data point; S36, based on the calculated connectivity, and comparing it with a preset threshold, identifying a time point when the connectivity is lower than the preset threshold; The method of calculating the weight by using the Gaussian function to calculate the Euclidean distance between two points in the neighborhood graph and constructing a weight matrix comprises the following steps: S331, define a Gaussian function and calculate the weight between two points; S332, for each pair of liquid level data at a time point in the neighborhood graph, calculating the Euclidean distance between each pair of liquid level data at a time point; S333, using a Gaussian function to calculate the weight between each pair of liquid level data at each time point according to the Euclidean distance; S334, filling all calculated weights into the weight matrix to construct the weight matrix; The weight formula between each pair of liquid level data at each time point is: ; In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between represents the natural exponential function; Indicates the liquid level data at a point in time i Liquid level data at time points j The Euclidean distance between represents the standard deviation of the Gaussian function; Represents the part of the Gaussian function that controls the speed of weight decay.
2. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 1 is characterized in that: The step of analyzing the collected liquid level data, calculating the liquid level difference within a preset time period, and setting an initial liquid level difference compensation value according to the liquid level difference comprises the following steps: S21, collecting liquid level data within a preset time period; S22, calculating the liquid level difference within the preset time period by comparing the liquid level readings at the beginning and end of the time period according to the liquid level data within the preset time period; S23, analyzing the trend of liquid level change according to the calculated liquid level difference, and determining the normal liquid level change range; S24, setting an initial liquid level difference compensation value according to the analysis results and the normal liquid level variation range; S25. Record the calculated liquid level difference and the set initial liquid level difference compensation value.
3. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 1 is characterized in that: The method of calculating the connectivity of each data point using the ADPP algorithm formula includes the following steps: S351, for each time point of the liquid level data in the multivariate time series data set, determine the liquid level data of all neighboring time points of the liquid level data at each time point in the neighborhood graph; S352, for the liquid level data at each time point, calculating the connectivity of the liquid level data at each time point; S353, recording the connectivity of the liquid level data at each time point, and comparing it with a preset threshold; The formula for calculating the connectivity of the liquid level data at each time point is: ; In the formula, Indicates the liquid level data at a point in time i Liquid level data at time points j The weight between k Represents the liquid level data at adjacent time points; Indicates the liquid level data at a point in time i Connectivity i , j All represent the liquid level data at a time point.
4. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 3 is characterized in that: The method of adjusting the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level change and the seasonal fluctuation of the historical liquid level data includes the following steps: S41, evaluating the change, amplitude and duration of the abnormal liquid level according to the time point when the connectivity is lower than the preset threshold; S42, comparing the abnormal liquid level change with the historical data, combining the seasonal fluctuations of the historical liquid level data, identifying whether there are common abnormalities under specific conditions, and determining whether the common abnormalities are in the initial liquid level difference compensation value; S43, if it is found that the abnormal liquid level changes are concentrated under specific conditions, the initial liquid level difference compensation value is adjusted under the specific conditions; S44, if the abnormal liquid level change is globally distributed, globally adjust the initial liquid level difference compensation value; S45. Re-perform abnormality detection using the adjusted liquid level difference compensation value, and record each adjusted liquid level difference compensation value and the corresponding abnormality detection result.
5. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 1 is characterized in that: The step of evaluating the adjusted liquid level difference compensation value and the abnormal detection result to determine whether there is leakage includes the following steps: S51, analyzing whether the abnormal detection result matches the known leakage event according to the liquid level difference compensation value after each adjustment and the corresponding abnormal detection result; S52, comparing the currently detected anomaly with the historical anomaly data in detail to evaluate whether there are repeated anomaly patterns and whether the anomaly patterns have been associated with leakage events in the past; S53, by comparing the abnormal detection results before and after the adjustment, determining whether false positives or false negatives are reduced; S54: Based on the evaluation results of steps S51 to S53, if the detected anomaly conforms to a known leakage pattern and is consistent with historical leakage data, it is determined that there is a sign of leakage.
6. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 1 is characterized in that: The multi-level early warning mechanism is set up. When it is determined that there are signs of leakage, the corresponding level of alarm is triggered according to the identification result, and corresponding emergency measures are taken, including the following steps: S61. Based on the severity of the leakage signs detected in real time, define multiple warning levels, and formulate emergency response measures for each warning level, wherein the multiple warning levels include a low warning level, a medium warning level, and a high warning level; S62, using the obtained real-time anomaly detection results, and classifying the anomaly detection results into corresponding warning levels according to severity; S63, matching the corresponding leakage pattern according to the type and characteristics of the real-time anomaly; S64. If the abnormal detection result is identified and classified as a specific warning level, the warning signal corresponding to the corresponding level and the pre-set emergency response measures are automatically triggered; S65. Implement emergency response measures that match the triggered warning level.
7. The abnormal warning method for hydraulic lubrication system for leakage detection according to claim 6 is characterized in that: If the abnormal detection result is identified and classified as a specific warning level, the warning signal corresponding to the corresponding level and the pre-set emergency response measures are automatically triggered, including the following steps: S641. Configure a unique alarm signal for each warning level; S642. Based on the leakage pattern matching result, define a triggered warning level for each type and characteristic of anomaly; S643. Establish an automated response mechanism to transmit warning signals and execute emergency response measures when warnings are triggered; S644. Collect feedback information when implementing emergency response measures, and use the feedback information to evaluate the effectiveness of the response measures and identify problems in implementation.
8. A hydraulic lubrication system abnormality warning system for leak detection, used to implement the hydraulic lubrication system abnormality warning method for leak detection according to any one of claims 1 to 7, characterized in that: The system includes: a liquid level data acquisition module, a liquid level data analysis module, an abnormality detection adjustment module, a liquid level difference compensation adjustment module, a leakage sign assessment module and a multi-level early warning mechanism module; The liquid level data acquisition module is connected to the abnormality detection adjustment module through the liquid level data analysis module, the abnormality detection adjustment module is connected to the leakage sign evaluation module through the liquid level difference compensation adjustment module, and the leakage sign evaluation module is connected to the multi-level early warning mechanism module; Liquid level data acquisition module, used to collect the liquid level data of the oil tank in real time using the liquid level sensor; The liquid level data analysis module is used to analyze the collected liquid level data, calculate the liquid level difference within a preset time period, and set an initial liquid level difference compensation value according to the liquid level difference; An abnormality detection adjustment module is used to adjust the original liquid level difference using the initial liquid level difference compensation value according to the abnormality detection algorithm, and identify abnormal liquid level changes that deviate from the normal range; A liquid level difference compensation adjustment module is used to adjust the initial liquid level difference compensation value according to the analysis results of the identified abnormal liquid level changes and seasonal fluctuations of historical liquid level data; A leakage sign evaluation module is used to evaluate the adjusted liquid level difference compensation value and abnormal detection results to determine whether there is leakage sign; The multi-level early warning mechanism module is used to set up a multi-level early warning mechanism. When it is determined that there are signs of leakage, the corresponding level of alarm is triggered according to the identification results, and corresponding emergency measures are taken.
Citation Information
Patent Citations
Abnormal leakage detection system of hydraulic system
CN112879389A
Water cooling system leakage monitoring method
CN115046704A
Monitoring method and system suitable for hydraulic oil leakage, equipment and medium
CN116464691A
Early warning method for detecting leakage of hydraulic lubricating system
CN117848596A
