Intelligent monitoring and fault warning method for municipal drainage system

Abstract

The invention relates to the technical field of electronic digital data processing, and in particular to an intelligent monitoring and fault early warning method for a municipal drainage system, comprising: setting a plurality of water level monitoring points based on the municipal drainage system, collecting actual water level data, and obtaining historical water level data of each water level monitoring point; determining the time series reliability of the historical water level data, analyzing the water level changes of the water level monitoring points and the water level monitoring points on both sides, obtaining the spatiotemporal reliability in combination with the time series reliability, correcting the moving average part in the ARIMA model according to the spatiotemporal reliability, and obtaining the predicted water level data of the water level monitoring point based on the corrected ARIMA model; comparing the predicted water level data with the actual water level data, and performing fault early warning; that is, correcting the moving average part in the ARIMA model based on the spatiotemporal reliability, and then obtaining the predicted water level data, obtaining more real and reliable data through verification, and improving the accuracy of the predicted water level data, so as to make the fault early warning for the pipeline more accurate.

Description

Technical Field

[0001] The invention relates to the technical field of electrical digital data processing, and in particular to a method for intelligent monitoring and fault early warning of a municipal drainage system. Background Art

[0002] According to different drainage methods, municipal drainage systems can be divided into rainwater drainage systems, sewage drainage systems and combined drainage systems. Among them, the rainwater drainage system is mainly used to quickly drain precipitation water and reduce the risk of flood disasters in heavy rain and heavy rainfall. The system can effectively drain a large amount of rainwater and prevent water accumulation in urban areas and roads, thereby ensuring the normal life of residents and smooth urban traffic. Therefore, by monitoring the operating status of rainwater drainage pipes, possible blockages or damage to the pipes can be discovered early, so that timely maintenance and repair can be carried out to enhance the city's disaster response capabilities.

[0003] In order to achieve effective monitoring, water level monitoring points are usually set up in rainwater drainage pipes, and the water level changes at the current moment are predicted by analyzing historical data. Water level data usually have time series characteristics, that is, there are certain patterns and trends in water level changes over time; the Auto-Regressive Moving Average Model (ARIMA) is usually used for analysis and prediction, which can capture the autocorrelation and trend in the data; however, this prediction method relies on the reliability of historical data. If the quality of historical data is poor, the prediction results may have large deviations, which will lead to inaccurate warnings; and water level data in actual monitoring usually changes dynamically, which makes it even more necessary to ensure the data quality of historical data to improve the accuracy of warnings. Summary of the invention

[0004] In order to solve the technical problem that the water level prediction using the ARIMA model is highly dependent on historical data, and the quality of historical data cannot be guaranteed, resulting in inaccurate warning, the purpose of the present invention is to provide a municipal drainage system intelligent monitoring and fault warning method, and the technical solution adopted is as follows:

[0005] Set up several water level monitoring points based on the municipal drainage system, collect actual water level data, and obtain historical water level data for each water level monitoring point;

[0006] Determine the time series reliability of historical water level data, analyze the water level changes at the water level monitoring point and the water level monitoring points on both sides, combine the time series reliability to obtain the spatiotemporal reliability, correct the moving average part of the ARIMA model according to the spatiotemporal reliability, and obtain the predicted water level data of the water level monitoring point based on the corrected ARIMA model;

[0007] Compare the predicted water level data with the actual water level data and issue fault warnings.

[0008] Preferably, determining the time series reliability of historical water level data includes:

[0009] Define any water level data in the historical water level data as target water level data, generate a time-water level function through the historical water level data, and obtain all historical water level data with the same water level as the target water level data in the historical water level data based on the time-water level function;

[0010] Determine the consistency of the target water level data with the surrounding trends of each historical water level data at the same water level, and obtain the temporal reliability of the target water level data in the time series.

[0011] Preferably, the consistency of the target water level data with the surrounding trend of any historical water level data at the same water level is determined, and the corresponding calculation formula is:

[0012]

[0013] in, Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; Indicates the time corresponding to the target water level data; Indicates any historical water level data The corresponding time; Indicates the general drainage time; represents the time-water level function; Indicates that the time-water level function is applied to any historical data The direction of translation .

[0014] Preferably, the time series reliability of the target water level data in the time series is obtained, and the corresponding calculation formula is:

[0015]

[0016] in, Represents the temporal reliability of target water level data in time series; Indicates selecting the maximum value; Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the upstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the historical water level data of the water level monitoring point on the left or right upstream of the municipal drainage system The historical water level data monitored at the corresponding time; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the downstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the water level data of the left or right water level monitoring point downstream in the municipal drainage system at any historical water level data. The historical water level data monitored at the corresponding time; Indicates the total number of all historical water level data that have the same water level as the target water level data.

[0017] Preferably, the water level changes at the water level monitoring point and the water level monitoring points on both sides are analyzed, and the spatiotemporal reliability is obtained by combining the time series reliability, including:

[0018] Arbitrarily obtain the offset time, combine the time-water level function to determine the offset time probability of the water level change between the water level monitoring point and the left and right water level monitoring points in the municipal drainage system, determine the actual offset time, and calculate the elevation of the water level monitoring point in the event of silt accumulation;

[0019] Based on the actual offset time and the elevation of the water level monitoring point due to specific circumstances, the target water level data and the drainage similarity of the water level monitoring points on both sides are analyzed, and the spatiotemporal reliability is obtained through the time series reliability.

[0020] Preferably, determining the actual offset time includes calculating the offset time probability of a water level change between any water level monitoring point and the left water level monitoring point, and the corresponding calculation formula is:

[0021] Get any offset time , represents the time difference between the response of the water level monitoring point and the water level monitoring points on both sides;

[0022]

[0023] in, Indicates the offset time between the water level monitoring point and the water level monitoring point on the left The probability of the offset time; Indicates An exponential function with base ; Indicates the time when water level monitoring point collection starts; Indicates the time when the water level monitoring point collection ends; represents the time-water level function; Indicates the time when the water level monitoring point on the left was generated - the water level function is shifted to the right by the time The time-water level function formed later;

[0024] Similarly, the probability of the offset time of water level change between each water level monitoring point and the water level monitoring points on both sides is obtained, and they are sorted from large to small. The offset time corresponding to the largest offset time probability is the actual offset time.

[0025] Calculate the elevation of the water level monitoring point under certain circumstances. The corresponding calculation formula is:

[0026]

[0027] in, Indicates that the water level monitoring point is raised due to specific circumstances; Indicates the historical water level data that appears most frequently in the historical water level data; Indicates selecting the maximum value; Indicates the historical water level data The number of times the historical water level data appears; Indicates that there are a total of kind;

[0028] Analyze the drainage similarity between the target water level data and the water level monitoring point on the left, and the corresponding calculation formula is:

[0029]

[0030] in, Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the water level corresponding to the target water level data; Indicates that the water level monitoring point corresponding to the target water level data has risen under certain circumstances; Indicates the water level corresponding to the target water level data after it is shifted to the left according to the actual shift time; It means that after the target water level data is shifted to the left according to the actual shift time, the corresponding water level monitoring point appears to be elevated under specific circumstances; similarly, the drainage similarity between the target water level data and the water level monitoring point on the right is analyzed.

[0031] Preferably, the spatiotemporal reliability is obtained, and the corresponding calculation formula is:

[0032]

[0033] in, Indicates the spatiotemporal reliability of target water level data; Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the drainage similarity between the target water level data and the water level monitoring point on the right; Indicates the time series reliability of target water level data.

[0034] Preferably, the moving average part in the ARIMA model is modified according to the spatiotemporal reliability, and the corresponding calculation formula is:

[0035]

[0036] in, represents the modified moving average part; represents a constant term; express Timing error; represents the number of moving average items; express Before the moment time; Indicates The spatiotemporal reliability of historical water level data at all times; express Timing error; Represents the error parameter.

[0037] Preferably, the predicted water level data and the actual water level data are compared to provide a fault warning, including:

[0038] Compare the predicted water level data with the actual water level data. If the predicted water level data and the actual water level data are different, determine the possibility of potential risks at the water level monitoring point corresponding to the predicted water level data;

[0039] Set a judgment threshold. If the possibility of potential risk at the water level monitoring point corresponding to the predicted water level data is greater than the judgment threshold, a fault warning is issued.

[0040] Preferably, the possibility of a potential risk existing at a water level monitoring point corresponding to the predicted water level data is determined, and the corresponding calculation formula is:

[0041]

[0042] in, Indicates that the water level monitoring point corresponding to the predicted water level data is There is always the possibility of potential risks; Indicates the general drainage time; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The actual water level data at the moment and the value is not 0; express The difference rate between the predicted water level data and the actual water level data at each moment; Indicates the predicted water level data corresponding to The actual water level data at the moment and the value is not 0; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The difference rate between the predicted water level data and the actual water level data at each moment.

[0043] The present invention has the following beneficial effects:

[0044] The technical solution of the present application obtains the time series reliability of the historical water level data by performing feature analysis to determine whether there is trend consistency between the historical water level data, and then determines the actual offset time by combining the water level changes due to inertia at the monitoring points on the left and right sides of the municipal drainage system, and further performs feature analysis in combination with the characteristics of specific situations such as silt accumulation to obtain the spatiotemporal reliability of all historical water level data, corrects the moving average part in the ARIMA model based on the spatiotemporal reliability, and then uses the corrected ARIMA model to obtain predicted water level data, further verifies the predicted water level data, obtains more real and reliable data, and improves the accuracy of the predicted water level data, thereby making the fault warning of the pipeline more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A flowchart of the steps of a method for intelligent monitoring and fault warning of a municipal drainage system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent monitoring and fault warning method of a municipal drainage system proposed by the present invention, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0049] The specific scheme of the municipal drainage system intelligent monitoring and fault warning method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a flowchart of the steps of a method for intelligent monitoring and fault warning of a municipal drainage system provided by an embodiment of the present invention, the method comprising:

[0051] Step S1: setting up a number of water level monitoring points based on the municipal drainage system, collecting actual water level data, and obtaining historical water level data of each water level monitoring point;

[0052] Step S2: determine the time series reliability of historical water level data, analyze the water level changes of the water level monitoring point and the water level monitoring points on both sides, obtain the spatiotemporal reliability in combination with the time series reliability, correct the moving average part of the ARIMA model according to the spatiotemporal reliability, and obtain the predicted water level data of the water level monitoring point based on the corrected ARIMA model;

[0053] Step S3: Compare the predicted water level data with the actual water level data and issue a fault warning.

[0054] For better explanation, the municipal drainage system includes a rainwater drainage system, a sewage drainage system and a combined drainage system. In this embodiment, the analysis is mainly based on the rainwater drainage system. The rainwater drainage system plays a vital role in urban drainage management. Heavy rainfall and rainstorms can cause waterlogging in urban areas, seriously affecting traffic, residents' lives and infrastructure operations. Timely and accurate warnings of possible blockages, ruptures or other failures in pipelines can help relevant departments prepare emergency plans in advance and reduce the risk of flood disasters. In the rainwater drainage system, water level monitoring points are usually set to monitor the operating status of the rainwater drainage system, and the ARIMA model is used to obtain future predicted water level data based on actual water level data. The ARIMA model consists of three main parts: an autoregressive (AR) part, which represents the relationship between the current value of the time series and its values ​​at previous time points; a difference (I) part, which eliminates the trend in the time series through a difference operation to make it stable; and a moving average (MA) part, which describes the influence of noise in the time series through a weighted average of the previous prediction errors.

[0055] However, the traditional ARIMA model predicts water level data by assuming that all historical water level data are reliable. In the face of inconsistent data quality, this may lead to deviations in prediction results. In actual situations, historical water level data may be missing or abnormal historical water level data may appear due to various factors such as equipment failure and network problems, resulting in inconsistent quality of different water level monitoring points, affecting the fitting effect of the ARIMA model and thus affecting the results of predicting water level data. In addition, the existing corrections to the ARIMA model are usually to average the weights of the moving average part, but in actual applications, water level data usually show periodic changes and may be affected by various factors such as season, weather, and precipitation. The average distribution weight is not completely valid, affecting the quality of water level data and thus affecting the predicted water level data.

[0056] Based on the above situation, the technical solution of the present application is to modify the ARIMA model and redistribute the weight of the moving average part to obtain accurate predicted water level data.

[0057] To illustrate, in step S1, a number of water level monitoring points are set up based on the municipal drainage system, actual water level data are collected, and historical water level data of each water level monitoring point is obtained.

[0058] Specifically, in each drainage pipe of the rainwater drainage system in the municipal drainage system, a water level monitoring point is set every 1 km and a water level sensor is installed, wherein the water level sensors include but are not limited to ultrasonic water level sensors, pressure water level sensors and float water level sensors; the collection frequency of each water level sensor is set to 12 times / hour, that is, the actual water level data is collected every 5 minutes, the actual water level data of the water level sensor is obtained in real time, and the historical water level data saved by the water level sensor in the current water level monitoring point is obtained.

[0059] It can be understood that since the change of water level over time has certain patterns and trends, for each historical water level data, the trend formed by it and the surrounding historical water level data can definitely be found in the entire historical water level data. There is not necessarily only one such trend. Therefore, by analyzing each consistent trend, the time series reliability of the historical water level data can be determined.

[0060] Furthermore, in step S2, determining the time series reliability of historical water level data includes:

[0061] Step S211: define any water level data in the historical water level data as target water level data, generate a time-water level function through the historical water level data, and obtain all historical water level data with the same water level as the target water level data in the historical water level data based on the time-water level function.

[0062] To explain, for the historical water level data obtained from the water level monitoring point, each adjacent historical water level data is connected with a broken line to generate a time-water level function, and all historical water level data with the same water level as the target water level data are obtained from all historical water level data, that is, in the time-water level function, the horizontal axis is time, and the vertical axis is historical water level data. Any water level data in the historical water level data is defined as the target water level data, which is recorded as , get the target water level data All historical water level data of the same vertical axis data point (same water level) are recorded as .

[0063] Step S212: Determine the consistency of the target water level data with the surrounding trends of each historical water level data at the same water level, and obtain the time series reliability of the target water level data in the time series.

[0064] Explain and analyze target water level data The similarity between the target water level data and the historical water level data with the same water level obtains the consistency of the trend around the target water level data and any historical water level data at the same water level, wherein the general drainage time refers to the time required for an area, system or equipment to remove the accumulated water. In this embodiment, the general drainage time is the drainage time required for the rainwater drainage system to drain water in the past.

[0065] Furthermore, in step S212, the consistency of the target water level data with the surrounding trend of any historical water level data at the same water level is determined, and the corresponding calculation formula is:

[0066]

[0067] in, Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; Indicates the time corresponding to the target water level data; Indicates any historical water level data The corresponding time; Indicates the general drainage time; represents the time-water level function; Indicates that the time-water level function is applied to any historical data The direction of translation .

[0068] To explain, by determining the consistency of the trend around the target water level data and any historical water level data at the same water level, that is, after the time-water level function is translated, the difference between the time-water level function and the untranslated time-water level function is calculated. The smaller the difference, the higher the consistency with the surrounding trends; The difference refers to the difference between the historical water level data with the same water level as the target water level data after the entire time-water level function is translated. The difference between the neighborhoods, i.e., the quantization The degree of difference between historical water level data during this time period.

[0069] It can be explained that, based on the above-mentioned thinking logic, the trend consistency of all historical water level data with the same water level as the target water level data is obtained; in addition, if the target water level data has a high consistency in trend with any historical water level data with the same water level, this means that not only the trend of the target water level data currently analyzed is consistent with the trend of the water level monitoring point corresponding to any historical water level data with the same water level, but also the drainage of the surrounding water level monitoring points at the time corresponding to these two historical water level data needs to be consistent. Therefore, the trend consistency is used to determine the contribution of all historical water level data with the same water level as the target water level data to the time series reliability.

[0070] Furthermore, in step S212, the time series reliability of the target water level data in the time series is obtained, and the corresponding calculation formula is:

[0071]

[0072] in, Represents the temporal reliability of target water level data in time series; Indicates selecting the maximum value; Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the upstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the historical water level data of the water level monitoring point on the left or right upstream of the municipal drainage system The historical water level data monitored at the corresponding time; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the downstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the water level data of the left or right water level monitoring point downstream in the municipal drainage system at any historical water level data. The historical water level data monitored at the corresponding time; Indicates the total number of all historical water level data that have the same water level as the target water level data.

[0073] To explain, in this embodiment, the pipeline where the water level monitoring point is set is a unidirectional drainage pipe, and there are only two directions, namely the left and the right sides. Based on the actual situation, the left water level monitoring point or the right water level monitoring point is analyzed for the direction of the target water level data and the historical water level data with the same water level. It can be understood that based on the water level monitoring point corresponding to the target water level data, whether it is the left water level monitoring point or the right water level monitoring point, the time used for analysis at this time is consistent and has not changed, that is, for the same time, whether the historical water level data obtained on the left or right side is the same, so the significance of selecting the historical water level data of the left water level monitoring point or the right water level monitoring point is also the same; based on the actual drainage situation, it is explained that the water level in the current pipeline is evenly distributed in the analyzed section, and the historical water level data recorded by different water level monitoring points at the same time point will not have significant differences.

[0074] It can be explained that, in actual situations, historical water level data and target water level data corresponding to the highest trend consistency are usually selected for analysis to simplify the amount of data calculation; it can be understood that as long as the target water level data and the historical water level data with the same water level have the highest trend consistency, it can be proved that the historical water level data is reliable, that is, it can effectively represent the change pattern of the historical water level data; therefore, correction is made based on the highest trend consistency, the deviation of the target water level data is adjusted, and the time series reliability of the target water level data in the time series is obtained.

[0075] It can be understood that the drainage pipe is a continuous system, and the fluid inside the pipe has connectivity. When external factors, such as rainwater input, stimulate the drainage pipe, the response of the fluid shows inertial characteristics, indicating that the change of water level in the drainage pipe will not be immediately reflected in all water level monitoring points, but will gradually expand and stabilize over time. This makes the water level changes between the current water level monitoring point and the adjacent water level monitoring points show certain similarities, and this similarity has a time offset; specifically, the water level changes between the current water level monitoring point and its adjacent water level monitoring points often show similar changing trends after a certain time delay.

[0076] Therefore, in order to determine the spatiotemporal reliability, it is usually necessary to first determine the inertial offset time, that is, the time delay required for the water level response to propagate; then, the time offset is used to evaluate the water level drainage similarity between the current water level monitoring point and the adjacent water level monitoring points on both sides using the integration method; if factors such as silt accumulation occur near the current water level monitoring point, causing the pipeline water level to rise abnormally, the result produced by the integral calculation may be biased, so it is also necessary to consider the impact of the water level rise on the calculation results, and therefore it is necessary to correct this part to obtain the time-water level function, so as to correct the time series reliability through the spatial characteristics and obtain the spatiotemporal reliability.

[0077] Furthermore, in step S2, the water level changes at the water level monitoring point and the water level monitoring points on both sides are analyzed, and the spatiotemporal reliability is obtained by combining the time series reliability, including:

[0078] Step S221: arbitrarily obtain the offset time, and combine the time-water level function to determine the offset time probability of water level changes between the water level monitoring point and the left water level monitoring point and the right water level monitoring point in the municipal drainage system, determine the actual offset time, and calculate the elevation of the water level monitoring point caused by silt accumulation.

[0079] It should be noted that the offset time refers to the time delay between the water level change at the current water level monitoring point and the water level monitoring points on both sides.

[0080] Furthermore, in step S221, determining the actual offset time includes calculating the offset time probability of a water level change between any water level monitoring point and the left water level monitoring point, and the corresponding calculation formula is:

[0081] Get any offset time , represents the time difference between the response of the water level monitoring point and the water level monitoring points on both sides;

[0082]

[0083] in, Indicates the offset time between the water level monitoring point and the water level monitoring point on the left The probability of the offset time; Indicates An exponential function with base ; Indicates the time when water level monitoring point collection starts; Indicates the time when the water level monitoring point collection ends; represents the time-water level function; Indicates the time when the water level monitoring point on the left was generated - the water level function is shifted to the right by the time The time-water level function formed later;

[0084] Similarly, the probability of the offset time of water level change between each water level monitoring point and the water level monitoring points on both sides is obtained and sorted from large to small. The offset time corresponding to the largest offset time probability is the actual offset time.

[0085] Specifically, based on the above-mentioned thinking logic, it can be clearly seen that the current water level monitoring point and the adjacent water level monitoring points on both sides are similar in a certain time offset, that is, the historical water level data obtained by the left water level monitoring point is translated to the right by a certain unit, or the historical water level data obtained by the right water level monitoring point is translated to the left by a certain unit to coincide with the historical water level data obtained by the current water level monitoring point, and the offset time is arbitrarily obtained. , calculate the probability of the offset time between the current water level monitoring point and the water level monitoring point on the left or right, that is, at the offset time In the case of , confirm the integral similarity between the time-water level function without translation and the new time-water level function generated after translation. The higher the integral similarity, the more it can prove that any offset time obtained is It does cause a shift in historical water level data.

[0086] Based on the probability of the offset time of water level changes between each water level monitoring point and the water level monitoring point on the left and the water level monitoring point on the right, they are sorted from large to small, and the offset time corresponding to the largest offset time probability is the actual offset time, that is, the offset time corresponding to the highest integral similarity is prioritized as the actual offset time to determine the typical time of delay or advance of water level changes.

[0087] Calculate the elevation of the water level monitoring point under certain circumstances. The corresponding calculation formula is:

[0088]

[0089] in, Indicates that the water level monitoring point is raised due to specific circumstances; Indicates the historical water level data that appears most frequently in the historical water level data; Indicates selecting the maximum value; Indicates the historical water level data The number of times the historical water level data appears; Indicates that there are a total of kind.

[0090] It is explained that, based on the actual situation, the most historical water level data obtained are undrained water level data, which means that undrained water level data frequently appears in the historical water level data, which may represent a normal or extreme level, so the historical water level data with the most occurrences are selected as the undrained water level data; if the value of the elevation of the water level monitoring point due to a specific situation calculated based on the data is not equal to 0, it means that there is currently a situation of silt elevation, that is, sediment accumulation causes the water level data to rise; therefore, in the subsequent analysis process, it is necessary to make corrections based on this specific situation to prevent affecting the reliability of the historical water level data.

[0091] Step S222: Based on the actual offset time and the elevation of the water level monitoring point due to specific circumstances, the target water level data and the drainage similarity of the water level monitoring points on both sides are analyzed, and the spatiotemporal reliability is obtained through the temporal reliability.

[0092] To explain, time series reliability refers to the stability of the historical water level data of the municipal drainage system operating within a certain period of time, that is, it can ensure that the municipal drainage system works normally within this period of time without failure or performance degradation; spatiotemporal reliability refers to the consideration of both time factors and space factors, indicating the stability of the historical water level data of the municipal drainage system operating within a certain period of time and a certain space.

[0093] It can be understood that the drainage pipe is a continuous system, and the fluid inside it has fluidity. When stimulated by external factors, the response of the fluid shows inertial characteristics, indicating that the change of water level in the drainage pipe will not be immediately reflected in all water level monitoring points. Specifically, based on the same time, the change of historical water level data between different water level monitoring points is not obvious, but as time changes, that is, the entire time-water level function shifts, the water level between the current water level monitoring point and the adjacent water level monitoring point will change and show a certain drainage similarity; in other words, the change of historical water level data will not be immediately and synchronously reflected in all water level monitoring points, but as time goes by, the water level change trend between each water level monitoring point tends to be consistent, reflecting the continuity and dynamic characteristics of the fluid movement inside the drainage pipe.

[0094] Specifically, the target water level data and the drainage similarity of the water level monitoring point on the left are analyzed, and the corresponding calculation formula is:

[0095]

[0096] in, Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the water level corresponding to the target water level data; Indicates that the water level monitoring point corresponding to the target water level data has risen under certain circumstances; Indicates the water level corresponding to the target water level data after it is shifted to the left according to the actual shift time; It means that after the target water level data is shifted to the left according to the actual shift time, the corresponding water level monitoring point appears to be elevated under specific circumstances; similarly, the drainage similarity between the target water level data and the water level monitoring point on the right is analyzed.

[0097] It is explained that because externally stimulated rainwater has certain inertial characteristics when flowing in the drainage pipe, within the actual offset time, the actual drainage volume of the pipe, that is, the monitored historical water level data minus the water level changes caused by specific conditions such as silt accumulation, is relatively consistent, that is, each historical water level data in the time-water level function is offset in time according to the actual offset time, then each historical water level data after the offset presents a similar drainage pattern, which shows that despite the difference in time, the drainage process is essentially similar, so as to accurately analyze the drainage similarity between the water level monitoring point corresponding to the target water level data and the water level monitoring point on its left and the water level monitoring point on its right where water level changes occurred, so as to provide data support for the subsequent acquisition of accurate spatiotemporal reliability.

[0098] Furthermore, in step S222, the spatiotemporal reliability is obtained, and the corresponding calculation formula is:

[0099]

[0100] in, Indicates the spatiotemporal reliability of target water level data; Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the drainage similarity between the target water level data and the water level monitoring point on the right; Indicates the time series reliability of target water level data.

[0101] It should be noted that when evaluating the spatiotemporal reliability of historical water level data, the key is time series reliability, that is, finding trends similar to the target water level data, because the ARIMA model is based on historical trends for future predictions. If historical water level data with trends similar to the target water level data cannot be found, the basis for the prediction will become unstable; and because the flow of rainwater in pipes has certain inertial characteristics, it means that within the actual offset time, the drainage similarity is higher, and the higher the similarity, the higher the reliability of historical water level data for obtaining predicted water level data.

[0102] It can be understood that the moving average part is corrected in combination with the spatiotemporal reliability of the historical water level data obtained as mentioned above, and then the ARIMA model is corrected; so as to give greater weight to those historical water level data that are more real and reliable, while reducing the dependence on those historical water level data that may not be real and reliable, so as to more accurately reflect the actual situation of the historical water level data, improve the accuracy of obtaining predicted water level data, and ensure the credibility and practicality of the predicted water level data.

[0103] Furthermore, in step S2, the moving average part of the ARIMA model is corrected according to the spatiotemporal reliability, and the corresponding calculation formula is:

[0104]

[0105] in, represents the modified moving average part; represents a constant term; express Timing error; represents the number of moving average items; express Before the moment time; Indicates The spatiotemporal reliability of historical water level data at all times; express Timing error; Represents the error parameter.

[0106] To illustrate, for the MA part of the ARIMA model, that is, the moving average part, the original formula is:

[0107]

[0108] In practical applications, Before The power generation at each moment is expressed The moving average part of the moment, but Before The data at each moment may not be reliable because Before The changes of data points at each moment in time series are continuous. Some data points may have unreal changes, while other data points may be relatively reliable. Therefore, based on the actual situation, the MA part of the ARIMA model is corrected in combination with the spatiotemporal reliability, that is, the real and reliable data points are made larger, the unreal data points are made smaller, the dependence on relatively reliable data points is increased, and the dependence on unreliable data points is reduced, so as to obtain more realistic predicted water level data.

[0109] Specifically, the predicted water level data of the water level monitoring point is obtained based on the modified ARIMA model, that is, the modified moving average part is substituted into the ARIMA model, and the water level data is obtained by using the modified moving average part. The historical water level data at the time is used to predict Predicted water level data at the time.

[0110] Furthermore, in step S3, the predicted water level data and the actual water level data are compared to provide a fault warning, including:

[0111] Step S31: Compare the predicted water level data with the actual water level data. If the predicted water level data is different from the actual water level data, determine the possibility that there is a potential risk at the water level monitoring point corresponding to the predicted water level data.

[0112] Make a statement, based on the time corresponding to the current predicted water level data, obtain the actual water level data of the water level monitoring point at the corresponding time, and judge whether the predicted water level data and the actual water level data match. If the two results match, it means that the drainage pipe monitored by the water level monitoring point is operating normally; otherwise, if the two results do not match, judge whether there is a potential risk at the current water level monitoring point.

[0113] Further, in step S31, the possibility of a potential risk existing at the water level monitoring point corresponding to the predicted water level data is determined, and the corresponding calculation formula is:

[0114]

[0115] in, Indicates that the water level monitoring point corresponding to the predicted water level data is There is always the possibility of potential risks; Indicates the general drainage time; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The actual water level data at the moment and the value is not 0; express The difference rate between the predicted water level data and the actual water level data at each moment; Indicates the predicted water level data corresponding to The actual water level data at the moment and the value is not 0; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The difference rate between the predicted water level data and the actual water level data at each moment.

[0116] It is explained that if the historical water level data is the healthy and normal water level data of the pipes in the municipal drainage system, then the obtained predicted water level data is also healthy water level data, which matches the actual water level data; on the contrary, the predicted water level data does not match the actual water level data, that is, data error occurs, indicating that there is a difference rate, The higher the value, the more likely it is that the drainage pipe has potential risks and the difference will continue. The higher it is, the greater the likelihood that there is risk.

[0117] Step S32: Setting a judgment threshold, if the possibility of a potential risk existing at the water level monitoring point corresponding to the predicted water level data is greater than the judgment threshold, a fault warning is issued.

[0118] As an optional implementation, in this embodiment, the judgment threshold is 0.7.

[0119] If the predicted water level data corresponds to the water level monitoring point at If the probability of potential risk at all times is greater than 0.7, a risk warning will be issued for the rainwater drainage pipe water level monitoring point to promptly discover and take necessary maintenance and repair measures for possible problems, so as to prevent and mitigate the impact of possible disasters in advance, enhance the city's disaster resistance and overall emergency response capabilities, effectively reduce losses caused by heavy rains or floods, ensure the stable operation of urban infrastructure, and protect the lives and property of citizens.

[0120] It can be understood that the technical solution of the present application obtains the time series reliability of the historical water level data by performing feature analysis to determine whether there is trend consistency between the historical water level data, and then determines the actual offset time by the water level changes due to inertia at the monitoring points on the left and right sides of the municipal drainage system, and further performs feature analysis in combination with the characteristics of specific situations such as silt accumulation to obtain the spatiotemporal reliability of all historical water level data, and corrects the moving average part in the ARIMA model based on the spatiotemporal reliability, and then uses the corrected ARIMA model to obtain predicted water level data, and further verifies the predicted water level data to obtain more real and reliable data, thereby improving the accuracy of the predicted water level data and providing more accurate warning of pipeline failures.

[0121] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0122] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for intelligent monitoring and fault warning of a municipal drainage system, characterized in that: The method comprises: Set up several water level monitoring points based on the municipal drainage system, collect actual water level data, and obtain historical water level data for each water level monitoring point; Determine the time series reliability of historical water level data, analyze the water level changes at the water level monitoring point and the water level monitoring points on both sides, combine the time series reliability to obtain the spatiotemporal reliability, correct the moving average part of the ARIMA model according to the spatiotemporal reliability, and obtain the predicted water level data of the water level monitoring point based on the corrected ARIMA model; Compare the predicted water level data with the actual water level data to provide fault warning; Determine the time series reliability of historical water level data, including: Define any water level data in the historical water level data as target water level data, generate a time-water level function through the historical water level data, and obtain all historical water level data with the same water level as the target water level data in the historical water level data based on the time-water level function; Determine the consistency of the target water level data with the surrounding trends of each historical water level data at the same water level, and obtain the temporal reliability of the target water level data in the time series; Analyze the water level changes at the water level monitoring point and the water level monitoring points on both sides, and combine the time series reliability to obtain the spatiotemporal reliability, including: Arbitrarily obtain the offset time, combine the time-water level function to determine the offset time probability of the water level change between the water level monitoring point and the left and right water level monitoring points in the municipal drainage system, determine the actual offset time, and calculate the elevation of the water level monitoring point in the event of silt accumulation; Based on the actual offset time and the elevation of the water level monitoring point due to specific circumstances, the target water level data and the drainage similarity of the water level monitoring points on both sides are analyzed, and the spatiotemporal reliability is obtained through the time series reliability.

2. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 1, characterized in that: Determine the consistency of the target water level data with the surrounding trend of any historical water level data at the same water level. The corresponding calculation formula is: in, Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; Indicates the time corresponding to the target water level data; Indicates any historical water level data The corresponding time; Indicates the general drainage time; represents the time-water level function; Indicates that the time-water level function is applied to any historical data The direction of translation .

3. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 2, characterized in that: The time series reliability of the target water level data in the time series is obtained, and the corresponding calculation formula is: in, Represents the temporal reliability of target water level data in time series; Indicates selecting the maximum value; Indicates the target water level data and any historical water level data at the same water level consistency with surrounding trends; Indicates An exponential function with base ; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the upstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the historical water level data of the water level monitoring point on the left or right upstream of the municipal drainage system The historical water level data monitored at the corresponding time; It indicates the historical water level data monitored by the water level monitoring point on the left or right side of the downstream in the municipal drainage system at the time corresponding to the target water level data; Indicates the water level data of the left or right water level monitoring point downstream in the municipal drainage system at any historical water level data. The historical water level data monitored at the corresponding time; Indicates the total number of all historical water level data that have the same water level as the target water level data.

4. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 1, characterized in that: Determine the actual offset time, including calculating the offset time probability of water level change between any water level monitoring point and the water level monitoring point on the left. The corresponding calculation formula is: Get any offset time , represents the time difference between the response of the water level monitoring point and the water level monitoring points on both sides; in, Indicates the offset time between the water level monitoring point and the water level monitoring point on the left The probability of the offset time; Indicates An exponential function with base ; Indicates the time when water level monitoring point collection starts; Indicates the time when the water level monitoring point collection ends; represents the time-water level function; Indicates the time when the water level monitoring point on the left was generated - the water level function is shifted to the right by the time The time-water level function formed later; Similarly, the probability of the offset time of water level change between each water level monitoring point and the water level monitoring points on both sides is obtained, and sorted from large to small. The offset time corresponding to the largest offset time probability is the actual offset time; Calculate the elevation of the water level monitoring point under certain circumstances. The corresponding calculation formula is: in, Indicates that the water level monitoring point is raised due to specific circumstances; Indicates the historical water level data that appears most frequently in the historical water level data; Indicates selecting the maximum value; Indicates the historical water level data The number of times the historical water level data appears; Indicates that there are a total of kind; Analyze the drainage similarity between the target water level data and the water level monitoring point on the left, and the corresponding calculation formula is: in, Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the water level corresponding to the target water level data; Indicates that the water level monitoring point corresponding to the target water level data has risen under certain circumstances; Indicates the water level corresponding to the target water level data after it is shifted to the left according to the actual shift time; It means that after the target water level data is shifted to the left according to the actual shift time, the corresponding water level monitoring point appears to be elevated under specific circumstances; similarly, the drainage similarity between the target water level data and the water level monitoring point on the right is analyzed.

5. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 4, characterized in that: The corresponding calculation formula for obtaining spatiotemporal reliability is: in, Indicates the spatiotemporal reliability of target water level data; Indicates the drainage similarity between the target water level data and the water level monitoring point on the left; Indicates the drainage similarity between the target water level data and the water level monitoring point on the right; Indicates the time series reliability of target water level data.

6. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 1, characterized in that: According to the spatiotemporal reliability correction of the moving average part of the ARIMA model, the corresponding calculation formula is: in, represents the modified moving average part; represents a constant term; express Timing error; represents the number of moving average items; express Before the moment time; Indicates The spatiotemporal reliability of historical water level data at all times; express Timing error; Represents the error parameter.

7. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 6, characterized in that: Compare the predicted water level data with the actual water level data to provide fault warnings, including: Compare the predicted water level data with the actual water level data. If the predicted water level data and the actual water level data are different, determine the possibility of potential risks at the water level monitoring point corresponding to the predicted water level data; Set a judgment threshold. If the possibility of potential risk at the water level monitoring point corresponding to the predicted water level data is greater than the judgment threshold, a fault warning is issued.

8. A municipal drainage system intelligent monitoring and fault warning method as claimed in claim 7, characterized in that: Determine the possibility of potential risk at the water level monitoring point corresponding to the predicted water level data. The corresponding calculation formula is: in, Indicates that the water level monitoring point corresponding to the predicted water level data is There is always the possibility of potential risks; Indicates the general drainage time; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The actual water level data at the moment and the value is not 0; express The difference rate between the predicted water level data and the actual water level data at each moment; Indicates the predicted water level data corresponding to The actual water level data at the moment and the value is not 0; Indicates the value obtained using the modified ARIMA model Predicted water level data at each moment; express The difference rate between the predicted water level data and the actual water level data at each moment.

Images