A method and system for monitoring the safety of underground spaces

Abstract

The present application relates to the field of data processing, more particularly, the present application relates to a kind of safety monitoring method and system of underground space, method includes: obtaining historical pressure data at pipe interface, extract the same time corresponding data sequence in preset period range and fit hyperplane;Based on the vertical distance of pressure data and hyperplane, calculate the periodicity index of each time, obtain the data of real-time pressure data before pre-set period, and calculate the importance of each time in pre-set period;According to periodicity index and importance, obtain the weight value of data in pre-set period, use weight value as weight in weighted average algorithm to predict residential water demand, use the difference between predicted value and actual value to carry out early warning.The present application can more accurately identify the real abnormal situation by comprehensively considering importance and periodicity index, reduce false alarm caused by normal periodic fluctuation, improve the accuracy and reliability of anomaly detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing. More particularly, the present application relates to a safety monitoring method and system for underground space. BACKGROUND

[0002] In the rapid development of urbanization process, as a valuable urban resource, the safety and stability of underground space has become the focus of urban planning and management. As the core infrastructure of underground space, water supply pipeline bears the heavy responsibility of providing residents with domestic water. The actual water supply main pipeline is usually laid in the underground pipe gallery. Due to the cramped space and poor ventilation inside the pipe gallery, the operating personnel often need to conduct regular inspection and maintenance, which cannot realize real-time monitoring of potential hazards of underground pipeline. If the pipeline seeps or leaks and is not treated in time, it will pose a potential threat to the overall stability of the underground space structure, and further exacerbate unsafe accidents. Therefore, it is necessary to monitor the internal data changes of the underground pipeline in real time, timely detect abnormalities and warn potential risks, and ensure the safety of the underground space structure.

[0003] The existing Chinese patent application file with publication number CN117252042A discloses a city underground space comprehensive bearing capacity evaluation system and method, which includes a foundation bearing capacity detector, an edge computing gateway analysis module and a bearing capacity evaluation module; the foundation bearing capacity detector is loaded with a sensor monitoring unit. The system collects relevant environmental data, and compares and judges the collected environmental data and performs BIM three-dimensional modeling analysis. When the compression modulus or liquidity index exceeds the preset safety limit value, the bearing capacity is evaluated, and when the preset safety limit value is not exceeded, the BIM three-dimensional modeling analysis of the changes in the underground environment of the building is performed.

[0004] The application file performs BIM three-dimensional modeling analysis on environmental data to assist workers in timely discovering local environmental changes and potential safety risks caused by construction, provides a basis for construction personnel to optimize construction process, enhances the timeliness of construction process supervision, and ensures construction safety. However, at present, due to the large demand for residential water in the morning and evening peak periods, the pressure of the pipeline will be affected by the fluctuation of water consumption. This fluctuation often shows periodic normal fluctuations. Relying only on local data and ignoring this normal periodic fluctuation will lead to misjudgment of the data by the pressure monitoring system, making it difficult to accurately identify potential risks of underground pipe gallery, and thus threatening the safety of underground space. SUMMARY

[0005] To solve the problem that relying only on local data and ignoring the normal periodic fluctuation will lead to misjudgment of the data by the pressure monitoring system due to the large demand for residential water in the morning and evening peak periods, the pressure of the pipeline will be affected by the fluctuation of water consumption, the present application provides solutions in the following aspects.

[0006] In a first aspect, a safety monitoring method for underground space includes: obtaining historical pressure data at a pipeline interface, extracting data sequences at the same time within a preset period range from the historical pressure data and fitting a hyperplane; calculating a periodicity index at each time according to the vertical distance of the pressure data corresponding to the arbitrary time from the hyperplane, obtaining data within a preset period before real-time pressure data, and calculating the importance of each time within the preset period; calculating the weight value of the pressure data within the preset period according to the periodicity index and the importance, using the weighted average algorithm, using the weight value as the weight of the pressure data, and weighting to predict the demand for residential water use, and using the difference between the predicted value and the actual value for early warning.

[0007] The effect is that by combining the periodicity analysis of historical pressure data and the importance evaluation of real-time data, the system can more accurately predict the demand for residential water use, improve the accuracy and reliability of the prediction, reduce false alarms caused by normal periodic fluctuations, and ensure that the system can accurately predict even in complex water use patterns; by fitting the hyperplane and calculating the periodicity index, the system can adapt to different water use patterns and periodic changes. Even during peak water use periods, the system can assess whether the current time pressure data conforms to the expected periodic pattern through the hyperplane fitted from historical data, thereby avoiding unnecessary alarms caused by normal fluctuations; by calculating the importance and weight value of each time within the preset period, the system can monitor the changes in pressure data in real time and use the difference between the predicted value and the actual value for early warning. This real-time monitoring and early warning mechanism can timely detect potential abnormalities such as pipeline leaks or pressure anomalies, so that timely maintenance measures can be taken.

[0008] Preferably, extracting data sequences at the same time within a preset period range from the historical pressure data and fitting a hyperplane includes:

[0009] Taking any time as the target time, traversing the pressure data at the same target time within a week of history, and constructing a pressure data matrix, while constructing a time index matrix according to the time index, fitting the pressure data matrix and the time index matrix using the least squares method to obtain the hyperplane.

[0010] The effect is that by traversing the pressure data at the same target time within a week of history, the pressure change trend and periodicity characteristics at that time can be captured. Using the least squares method to fit the hyperplane can establish a mathematical model that can more accurately predict the pressure data at the current target time; the hyperplane fitted by the least squares method not only considers the pressure data at the current target time, but also integrates the data at the same time within a week of history, and the fitting method based on historical data can better capture the long-term trend and periodic changes of the pressure data, thereby enhancing the generalization ability of the model.

[0011] Preferably, the periodicity indicator comprises:

[0012] Calculate the average of the vertical distance between the pressure data at the same time within the preset period range and the hyperplane, to obtain the average level of the preset period range;

[0013] Calculate the absolute deviation between the vertical distance of the pressure data at the current time and the hyperplane and the average level, and use the negative exponential function to exponentially attenuate the absolute deviation, to obtain the periodicity indicator at the current time.

[0014] Preferably, the importance degree comprises:

[0015] Taking any time as the target time, obtain the pressure data within the local period before the target time, calculate the sum of squares of differences between the pressure data at the first time and other times within the local period, calculate the average of the sum of squares of differences between the pressure data at other times within the local period and the average of the pressure data at each time within the local period except the first time, to obtain the standard deviation of the pressure data; normalize the ratio between the sum of squares of differences and the standard deviation, to obtain the importance degree of the first time within the local period before the target time.

[0016] The effect is that by evaluating the difference between the pressure data at each time and other times, and the fluctuation degree of the pressure data within the local period, the abnormal time can be more accurately identified, reducing misjudgment and ensuring that the system can timely discover and handle potential abnormal situations; it can effectively distinguish between normal fluctuations and abnormal fluctuations, avoiding unnecessary alarms due to normal fluctuations even during peak water consumption periods, ensuring that the system maintains stable performance in the face of complex water consumption patterns and periodic fluctuations, reducing maintenance costs and resource waste due to misjudgment.

[0017] Preferably, the importance degree further comprises:

[0018] Taking any time as the target time, obtain the pressure data within the local period before the target time, calculate the difference between the pressure data at the first time within the local period and the average of the pressure data at each time within the local period except the first time, normalize the ratio between the difference and the standard deviation of the pressure data at each time within the local period except the first time, to obtain the importance degree of the first time within the local period before the target time.

[0019] ​​​​​​​​Its effect lies in: by evaluating the difference between the pressure data of each moment and the mean value of other moments in the local period, combined with the ratio of standard deviation, the abnormal moment can be more accurately identified, and the system can timely discover and handle potential abnormal situations.

[0020] Preferably, the weight value comprises:

[0021] Taking any moment as a target moment, the ratio between the importance degree of the target moment and the periodicity index is calculated, and normalized processing is performed to obtain the weight value of the pressure data corresponding to the target moment.

[0022] Preferably, the weight value further comprises:

[0023] Taking any moment as a target moment, the ratio between the importance degree of the target moment and the periodicity index is calculated to obtain the abnormal degree of the target moment.

[0024] The product of the reciprocal of the difference between the pressure data of the target moment and the pressure data of the previous moment of the target moment and the abnormal degree is normalized and taken as the weight value of the pressure data corresponding to the target moment.

[0025] Its effect lies in: by calculating the ratio between the importance degree of the target moment and the periodicity index, the abnormality of the pressure data of the moment can be more accurately evaluated. The importance degree reflects the difference between the pressure data of the moment and other moments, and the periodicity index reflects the similarity between the pressure data of the moment and the historical same period. The weight value of each target moment is calculated, and the weight value reflects the relative importance of each moment in the overall monitoring, which helps to prioritize processing and attention to moments that may contain abnormal information, and improves the overall monitoring and maintenance efficiency.

[0026] Preferably, the warning by using the difference between the predicted value and the actual value comprises:

[0027] The absolute difference between the predicted value and the actual monitoring value of the current moment pressure data is normalized as the abnormal degree of the current moment;

[0028] In response to the abnormal degree of the current moment being greater than the warning coefficient, the warning information is sent to the monitoring center, prompting that there is a safety risk in the underground pipe gallery corresponding pipeline, which needs to be repaired in time, otherwise, the underground pipe gallery corresponding pipeline is safe.

[0029] The second aspect of the application is a safety monitoring system for underground space, comprising: a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, the safety monitoring method for underground space is realized.

[0030] The application has the following effects:

[0031] 1、The application can more accurately identify the real abnormal situation by comprehensively considering the importance and periodicity index. Not only the difference between the current time pressure data and other time is considered, but also the historical periodicity mode and the change trend of adjacent time are considered, reducing the false alarm caused by normal periodic fluctuation, improving the accuracy and reliability of abnormal detection.

[0032] 2、The application can effectively distinguish between normal fluctuation and abnormal fluctuation by introducing the periodicity index, so that during the peak period of water consumption, the system can also evaluate whether the current time pressure data conforms to the expected periodicity mode through the hyperplane fitted by the historical data, thereby avoiding unnecessary alarm triggered by normal fluctuation. BRIEF DESCRIPTION OF DRAWINGS

[0033] The above and other objects, features and advantages of the exemplary embodiments of the present application will be more apparent from the following detailed description read in conjunction with the accompanying drawings, in which several embodiments of the present application are shown by way of example, and wherein the same or corresponding elements refer to the same or corresponding parts wherein:

[0034] Figure 1 is a method flow chart of steps S1-S3 in a safety monitoring method of underground space according to an embodiment of the present application.

[0035] Figure 2 is a structure block diagram of a safety monitoring system of underground space according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are 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 are within the scope of protection of the present application.

[0037] The specific embodiments of the present application will be described in detail below with reference to the drawings.

[0038] Specific implementation scenario: by installing intelligent pressure monitors at key nodes of underground pipe gallery pipelines, such as interfaces, main pipe branches, etc., the pressure data at the pipe interface can be obtained in real time. These places are the key areas of pressure change in the pipeline system.

[0039] Reference Figure 1 A safety monitoring method of underground space includes steps S1-S3, as follows:

[0040] S1: Obtain historical pressure data at the pipe interface, extract the data sequence at the same moment within a preset period based on the historical pressure data, and fit a hyperplane.

[0041] In this embodiment, the preset sensor collects data once per minute to ensure that minute changes can be captured, making it easier to detect abnormalities in a timely manner.

[0042] Using any given moment as the target moment, we iterate through the pressure data for the same target moment within a week and construct a pressure data matrix. At the same time, we construct a time index matrix based on the time index. We then use the least squares method to fit the pressure data matrix and the time index matrix to obtain the hyperplane.

[0043] It should be noted that since the actual main water supply pipeline is deployed in the underground space of residential buildings, the internal pressure of the pipeline is affected by the daily water use patterns of residents. During the morning and evening peak hours, the pipeline pressure often fluctuates due to the surge in water consumption. This fluctuation is a normal phenomenon and has obvious regularity.

[0044] Therefore, if pressure data fluctuates at a certain moment, it is necessary to consider whether this fluctuation is a normal change caused by daily water usage patterns or a fluctuation caused by abnormal circumstances. Once the pressure data at a certain moment is closer to the historical pressure data of the same period to fit a historical trend curve representing normal behavior, and the smaller the difference between this distance and the average absolute distance of each pressure data point from the fitted curve in the historical period, it means that the change at that moment is consistent with historical normal behavior, and that is, the periodicity index is considered to be larger.

[0045] S2: Based on the vertical distance between the pressure data at any given time and the hyperplane, calculate the periodicity index at each time, obtain the data for a preset period before the real-time pressure data, and calculate the importance of each time within the preset period.

[0046] For example, for any given time, the corresponding pressure data for the same time within a preset historical period is obtained by iterating through historical timeframes, using the following iteration formula:

[0047] ;

[0048] in, This indicates the number of days to traverse within a preset period. An index representing the number of days within a preset period. Indicates the first At that moment, This represents the total number of minutes in a day. .

[0049] In this embodiment, the preset period is one week, but the implementer can adjust it according to the specific situation, such as two weeks or one month.

[0050] Calculate periodic indicators at any given time, including:

[0051] Taking any given moment as the target moment, calculate the vertical distance between the pressure data point at the target moment and the fitted hyperplane; calculate the mean of the vertical distances between the corresponding data points within a preset period range before the target moment and the fitted hyperplane; calculate the absolute difference between the vertical distance at the target moment and the mean of the vertical distances within the preset period range before the target moment, and use an exponential function to correct the absolute difference to obtain the periodic index of the target moment.

[0052] Specifically, the periodic indicators satisfy the following relationship:

[0053] ;

[0054] In the formula, Indicates the first Periodic indicators at any given time, Indicates the first The perpendicular distance between the data point and the hyperplane at each time step. Indicates the first A preset period range is defined before each time point. Indicates the first Before the first The perpendicular distance between the data point and the hyperplane at each time step. Represented by natural numbers An exponential function with base 0.

[0055] In other words, Indicates the first The perpendicular distance between the data point at time t and the hyperplane, the t-th time... The absolute difference between the mean vertical distances between the data points at the same time within a preset period before the nth time point and the hyperplane; the smaller this value, the more pronounced the nth time point. The more consistent the pressure data at a given moment is with the trend of change at the same moment in the historical cycle, the more consistent the pressure data at the current moment is with the expected periodic pattern, and the more likely that the moment is to be considered to have strong periodicity.

[0056] In addition, another embodiment includes:

[0057] Taking any given moment as the target moment, calculate the average absolute deviation between the pressure data at the target moment and the pressure data at the same moment within a previously preset periodic range. Then, use an exponential function to correct the average absolute deviation to obtain the periodic index for the target moment.

[0058] Specifically, the periodic indicators satisfy the following relationship:

[0059] ;

[0060] wherein, represents a periodicity index of the th moment, represents a preset period range before the th moment, represents pressure data of the th moment, represents pressure data of the th moment one day ago, represents pressure data of the th moment one day ago, represents an exponential function with a natural number

[0061] That is, reflects the mean difference between the pressure data of the target moment and the pressure data of the same moment within the preset period range before the target moment, the smaller the mean difference, the more consistent with the characteristics of the period change, the stronger the periodicity; the larger the periodicity index, the more consistent the change trend of the pressure data of the current moment and the pressure data of the same moment in history, that is, the current pressure change conforms to the historical periodicity mode; otherwise, the smaller the periodicity index, the less consistent the change trend of the pressure data of the current moment and the pressure data of the same moment in history, that is, the current moment may have an abnormal situation.

[0062] It should be noted that since the pipeline is exposed to a humid and poorly ventilated environment for a long time, the pipeline material may be slightly waterlogged or rusted due to interface corrosion, resulting in a rupture, and if the pipeline is not repaired in time and the water seepage is not cleaned up, long-term water seepage will affect the stability of the underground structure (problems such as reduction of concrete steel strength and foundation settlement). The actual pipeline water seepage or leakage problem will cause the water flow in the pipeline to become unstable, and when the water flow velocity changes, the pressure in the pipeline will also fluctuate, thereby forming a distribution characteristic that is significantly different from the normal stable pressure.

[0063] Therefore, the importance of each moment is evaluated by considering the change characteristics of the pressure data in the time sequence segment, if a moment shows data significantly different from other moments in the segment, it means that there may be a potential abnormal situation at that moment that causes the pressure data to fluctuate abnormally, and the more important the current moment is.

[0064] Specifically, the importance is obtained, including:

[0065] Taking any moment as a target moment, obtaining pressure data in a local segment before the target moment, calculating the sum of squares of differences between the pressure data of the th moment and other moments in the local segment, and calculating the sum of squares of differences between the pressure data of other moments in the local segment except the th moment and the pressure data of the The standard deviation of the pressure data is obtained by taking the mean of the sum of squared differences between the means of pressure data at each time point other than the target time point; the ratio of the sum of squared differences to the standard deviation is then normalized to obtain the standard deviation of the pressure data within the local time period before the target time point. The importance of each moment.

[0066] In this embodiment, the local time period is set to 10 data points, excluding the target time period.

[0067] Specifically, the importance level satisfies the following relationship:

[0068] ;

[0069] In the formula, Indicates the first time interval within a local time period at the current moment. The importance of each moment This indicates the number of times within a local time period at the current moment. This represents the time index value within a local time period at the current moment. Indicates the first time interval within a local time period at the current moment. The corresponding pressure data at each moment. Indicates the first time interval within a local time period at the current moment. The corresponding pressure data at each moment. Indicates the local time period excluding the first... The average pressure data for each of the other time points at that given time. This represents the standard normalization function.

[0070] In other words, Indicates the current local time period, the first... The sum of the squared differences between the pressure data at time 1 and the pressure data at every other time excluding that time is considered a positive sign. The larger this value, the stronger the positive sign. The greater the difference in pressure data at a given moment compared to other moments, the better.

[0071] Indicates the local time period excluding the first... The smaller the standard deviation of the pressure data at other times besides the first time point, the more stable the pressure data changes at other times within the local time period. In such a case, if the first time point... The greater the difference in pressure data at a given moment compared to other moments, the more it means that the first moment... If there are anomalies in the pressure data at a certain moment, then that moment is considered important.

[0072] In addition, another embodiment includes:

[0073] At any target moment, the pressure data in the local period before the target moment is obtained, the difference between the pressure data of the first moment in the local time and the average of the pressure data corresponding to each moment other than the first moment in the local period is calculated, the difference is normalized with the ratio of the standard deviation of the pressure data of each moment other than the first moment in the local period, and the importance of the first moment in the local period before the target moment is obtained.

[0074] Specifically, the importance satisfies the following relationship:

[0075] ;

[0076] Exemplarily, represents the importance of the first moment in the local period at the current moment, represents the pressure data corresponding to the first moment in the local period at the current moment, represents the average of the pressure data corresponding to each moment other than the first moment in the local period at the current moment, represents the standard deviation of the pressure data of each moment other than the first moment in the local period at the current moment, represents the standard normalization function. It should be noted that once the periodicity index of each moment is determined, it means that whether the change of the pressure data of each moment conforms to the fluctuation of the historical daily water consumption mode can be distinguished. If the data of a moment shows obvious fluctuation in the local period, and such fluctuation does not conform to the historical daily water consumption mode, it means that the data of this moment may be abnormal and needs to be paid attention to. Therefore, the weight value is dynamically adjusted in combination with the two indexes, so that those moments considered to be abnormal are given more attention in weighting, thereby enhancing the sensitivity to potential abnormal conditions and reducing the sensitivity to normal mode fluctuation data, so that the prediction value of the current moment is more consistent with the actual pipeline condition.

[0077] S3: According to the periodicity index and the importance, the weight value of the pressure data in the preset period is calculated, the weight value is used as the weight of the pressure data, and the weighted average algorithm is used for weighting to predict the water demand of residents, and the difference between the prediction value and the actual value is used for early warning.

[0078] The weight value includes:

[0079]

[0080] The weight value includes:

[0081] ​​​​​​​Taking any given moment as the target moment, calculate the ratio between the importance of the target moment and the periodic indicators, and perform normalization processing to obtain the weight value of the pressure data corresponding to the target moment.

[0082] Specifically, the weight values ​​satisfy the following relationship:

[0083] ;

[0084] In the formula, Indicates the first time interval within a local time period at the current moment. The weight value of the pressure data at each time point Indicates the first time interval within a local time period at the current moment. The importance of each moment Indicates the first time interval within a local time period at the current moment. Periodic indicators at any given time, This represents the standard normalization function.

[0085] In addition, another embodiment includes:

[0086] Taking any given moment as the target moment, calculate the ratio between the importance of the target moment and the periodic indicator to obtain the degree of anomaly at the target moment;

[0087] The product of the reciprocal of the difference between the pressure data at the target time and the pressure data at the time before the target time and the degree of anomaly, after normalization, is used as the weight value of the pressure data corresponding to the target time.

[0088] Specifically, the weight values ​​satisfy the following relationship:

[0089] ;

[0090] In the formula, Indicates the first time interval within a local time period at the current moment. The weight value of the pressure data at each time point Indicates the first time interval within a local time period at the current moment. The importance of each moment Indicates the first time interval within a local time period at the current moment. Periodic indicators at any given time, Indicates the first time interval within a local time period at the current moment. The corresponding pressure data at each moment. Indicates the first time interval within a local time period at the current moment. The corresponding pressure data at each moment. This represents the standard normalization function.

[0091] The absolute difference between the predicted and actual pressure values ​​at the current moment is normalized to determine the degree of anomaly at the current moment.

[0092] In response to the abnormal degree at the current time being greater than the early warning coefficient, early warning information is sent to the monitoring center, prompting that the underground pipe gallery corresponding pipeline has a security risk and needs to be repaired in time, otherwise, the underground pipe gallery corresponding pipeline is safe.

[0093] In this embodiment, the early warning coefficient is 0.75.

[0094] The application further provides a safety monitoring system of underground space. Figure 2 As shown in the figure, the system comprises a processor and a memory, and the memory stores computer program instructions, which realize the safety monitoring method of underground space according to the first aspect of the application when executed by the processor.

[0095] The system further comprises a communication bus and a communication interface and other components familiar to those skilled in the art, the settings and functions of which are known in the art, and thus will not be described here.

[0096] In the present application, the aforementioned memory can be any tangible medium containing or storing a program, which can be used or combined with an instruction execution system, device or apparatus. For example, the computer readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high bandwidth memory (HBM), hybrid memory cube (HMC) and the like, or any other medium that can be used to store the required information and can be accessed by an application, module or both. Any such computer storage medium can be part of the device or accessible or connectable to the device. Any application or module described in the present application can be implemented by computer readable / executable instructions stored or otherwise held by such computer readable medium.

[0097] In the description of the present application, the meaning of "a plurality of", "several" is at least two, for example, two, three or more, etc., unless otherwise explicitly specified.

[0098] While the specification has illustrated and described various embodiments of the application, it will be clear to those of ordinary skill in the art that various changes, modifications, and substitutions can be made thereto without departing from the spirit and scope of the application. It is understood that in the process of practicing the application, various alternatives, modifications, and equivalents can be employed.

Claims

1. A method of safety monitoring of an underground space, characterized in that, The method comprises the following steps: acquiring historical pressure data at the pipeline interface, extracting data sequences at the same time within a preset period range from the historical pressure data, and fitting a hyperplane, including: taking any time as a target time, traversing the pressure data at the same target time within a week, and constructing a pressure data matrix, and simultaneously constructing a time index matrix according to the time index, fitting the pressure data matrix and the time index matrix by using the least square method, and acquiring the hyperplane; calculating the periodicity index of each time according to the perpendicular distance of the pressure data corresponding to any time to the hyperplane, acquiring data in a preset period before the real-time pressure data, and calculating the importance of each time within the preset period; calculating the weight value of the pressure data within the preset period according to the periodicity index and the importance, using the weighted average algorithm, taking the weight value as the weight of the pressure data, and weighting to predict the residents' water demand, and using the difference between the predicted value and the actual value for early warning; The periodicity index comprises: calculating the average value of the perpendicular distance of the pressure data at the same time within a preset period range to the hyperplane, obtaining the average level of the preset period range; calculating the absolute deviation between the perpendicular distance of the pressure data at the current time to the hyperplane and the average level, and using a negative exponential function to exponentially attenuate the absolute deviation, obtaining the periodicity index at the current time.

2. The method of claim 1, wherein, The importance comprises: With any moment as a target moment, acquire pressure data in a local period before the target moment, calculate the sum of squares of differences between pressure data of the first moment and other moments in the local period, calculate the mean of the sum of squares of differences between pressure data of other moments except the first moment in the local period and the mean of pressure data of each moment except the first moment in the local period, and obtain the standard deviation of the pressure data; calculate the ratio between the sum of squares of differences and the standard deviation, and normalize the ratio to obtain the importance of the first moment in the local period before the target moment. ​​​​ 3. The method of claim 1, wherein the method further comprises: The importance further comprises: Taking any moment as a target moment, obtaining pressure data in a local period before the target moment, calculating a difference between pressure data of the first moment in the local period and a mean value of pressure data corresponding to each moment other than the first moment in the local period, normalizing the difference with a ratio of the pressure data of each moment other than the first moment in the local period to a standard deviation of the pressure data of each moment other than the first moment in the local period, to obtain an importance degree of the first moment in the local period before the target moment. ​​​​ 4. The method of claim 1, wherein, The weight value comprises: Taking any time as a target time, calculating the ratio between the importance and the periodicity index at the target time, and performing normalization processing to obtain the weight value of the pressure data corresponding to the target time.

5. The method of claim 1, wherein the method further comprises: The weight value further comprises: Taking any time as a target time, calculating the ratio between the importance and the periodicity index at the target time, obtaining the abnormality degree at the target time; The product of the reciprocal of the difference between the pressure data at the target time and the pressure data at the time before the target time and the abnormality degree is normalized and taken as the weight value of the pressure data corresponding to the target time.

6. The method of claim 1, wherein the method further comprises: The early warning using the difference between the predicted value and the actual value comprises: normalizing the absolute difference between the predicted value and the actual monitoring value of the current time pressure data as the abnormality degree of the current time; In response to the abnormality degree of the current time being greater than the early warning coefficient, sending early warning information to the monitoring center to prompt that the corresponding pipeline of the underground pipe gallery has a safety risk and needs to be repaired in time, otherwise, the corresponding pipeline of the underground pipe gallery is safe.

7. A safety monitoring system for an underground space, characterized in that The method comprises the following steps: a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, the method for monitoring the safety of underground space according to any one of claims 1-5 is realized.

Images