Multi-point pressure monitoring and early warning method and system for intelligent fire hydrant

Abstract

The invention discloses a multi-point pressure monitoring and early warning method and system for intelligent fire hydrants, and relates to the technical field of safety early warning of the intelligent fire hydrants.The method comprises the steps that when multi-point pressure monitoring is conducted on the intelligent fire hydrants, the pressure value of each intelligent fire hydrant is obtained through a pressure sensor; a microphone sensor is installed at a pipeline of the intelligent fire hydrant to obtain sound; the pressure condition of the intelligent fire hydrant is monitored by acquiring the specific position of the intelligent fire hydrant in the area of the farm area, the early warning threshold value and the abnormal sound type are set according to historical data, and whether the monitored real-time pressure value is an abnormal pressure value is predicted and judged so as to generate an abnormal monitoring signal; and early warning level analysis is carried out on the abnormal pressure value, whether the pressure drop is temporarily fluctuated or continuously deteriorated is distinguished, the change trend of the intelligent fire hydrant is effectively predicted to generate a monitoring early warning signal of the abnormal intelligent fire hydrant, and effective resource scheduling is carried out according to monitoring demand personnel of the department.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fire hydrant safety alarm technology, and in particular to a multi-point pressure monitoring and early warning method and system for intelligent fire hydrants. Background Art

[0002] Fire hydrants are a vital component of urban fire protection systems. Their pressure status is directly related to whether firefighting operations can be carried out promptly and effectively when a fire occurs. Smart fire hydrants are an intelligent upgrade of traditional fire hydrants, integrating advanced technologies such as the Internet of Things, sensors, and big data. Managers can view the status of all smart fire hydrants at any time through computers, mobile phones, and other terminal devices, enabling remote monitoring. They can also remotely control the valves of fire hydrants and perform emergency operations when necessary. With the rapid development of modern agriculture, large-scale farms are expanding in size, and various production facilities are becoming increasingly complex. Farms are typically home to numerous buildings, storage facilities, and crop planting areas, all of which have extremely high fire safety requirements. As a key component of the fire protection system, the proper functioning of fire hydrants is crucial to fire fighting.

[0003] In the operation of existing farms, fire hydrants are distributed over a wide area and at large intervals. When faced with such large-scale and diverse monitoring needs, fire hydrant pipes may leak or become clogged, leading to unstable pressure. A single pressure monitoring point cannot accurately reflect the pressure distribution of the entire farm fire hydrant system, making it difficult to comprehensively assess and warn of potential pressure hazards. The lack of an effective multi-point pressure monitoring and warning mechanism makes it difficult for farm managers to timely grasp the pressure status of each fire hydrant, often making them feel powerless.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide early warning by integrating multiple abnormal sound data, and to monitor and judge the abnormal pressure conditions of smart fire hydrants, distinguish whether the pressure drop is a temporary fluctuation or a continuous deterioration, so as to achieve a more accurate analysis of abnormal conditions.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a multi-point pressure monitoring and early warning method and system for intelligent fire hydrants, comprising the following steps:

[0007] Step 1: When performing multi-point pressure monitoring on smart fire hydrants, the pressure value of each smart fire hydrant is obtained through a pressure sensor. A microphone sensor is installed on the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants.

[0008] Step 2: Obtain historical data generated by smart fire hydrants from an online resource database. The historical data includes pressure and sound data of smart fire hydrants, and generate an early warning mechanism for monitoring smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets an early warning threshold based on historical data and uses the early warning threshold to determine abnormalities in real-time pressure values. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging.

[0009] Step 3: Monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and conduct abnormal monitoring judgment to generate abnormal monitoring signals;

[0010] Step 4: Establish a pressure value warning level based on the warning threshold, generate a warning level report, obtain the generated abnormal monitoring signal, and perform sound abnormality identification based on the pressure value of the abnormal smart fire hydrant to generate a sound abnormality type report. Based on the warning level report and the abnormality type report, obtain the monitoring warning signal of the abnormal smart fire hydrant;

[0011] Step 5: Obtain the monitoring and early warning signal of the abnormal smart fire hydrant, and output it to the department management end for intelligent personnel dispatch based on the monitoring demand personnel in the smart fire hydrant.

[0012] Furthermore, when the pressure value and sound of the smart fire hydrant are obtained in step one, the obtained pressure value and sound are sent as data to the data storage control end, the database storage control end stores and records them, and establishes a storage analysis table for each smart fire hydrant to record the pressure value and sound generated by the smart fire hydrant.

[0013] Furthermore, the specific process of judging abnormal conditions of real-time pressure values ​​through early warning thresholds is as follows:

[0014] Obtain the historical pressure data of the smart fire hydrants within the farm area, select the pressure value data within a period of time based on the historical pressure data, average the pressure data values ​​and remove the error, and obtain the calculated pressure value p: q is the number of pressure values ​​calculated, m = 1, 2, ..., q, p i is the pressure value of the i-th;

[0015] Furthermore, the abnormal sound recognition in step 2 specifically includes the following:

[0016] S200: Acquire historical sound data generated by the smart fire hydrant in the event of pipe blockage, valve damage, or pipe aging, and integrate the historical sound data as audio data to obtain an audio dataset;

[0017] A sound recognition model is built based on the audio dataset using a convolutional neural network. The sound recognition model consists of an input layer, a recognition layer, and an output layer. The audio dataset is used as input. The input layer classifies and labels the audio dataset, and performs error and noise removal preprocessing on the audio data to obtain the processed audio dataset.

[0018] S201: The recognition layer obtains the processed audio data set and extracts the frequency and time domain features of the audio data to obtain the extracted feature set. The frequency features are the basic frequency and main frequency of the generated sound, and the time domain features are the duration of the sound and the time amplitude change rate. The amplitude change rate is calculated by the time change of the audio data. The specific recognition process is as follows:

[0019] Setting related functions

[0020] According to the sampling frequency F S Calculate the main frequency F: m=1,

[0021] Where: R[T] is the autocorrelation function value of signal x[n], which describes the correlation of the signal itself when the delay is K, K is the delay amount, which is used to represent the comparison of the signal itself at different time offsets, n is the index of the discrete time series, N is the number of sample points of signal x[n], K0 is the delay corresponding to the first peak in the autocorrelation function except K=0, m is the frequency domain index, X[m] is the discrete Fourier transform result, which is the mth coefficient in the frequency domain, j is a virtual unit, and M is the index corresponding to the main frequency;

[0022] Calculate the duration T of the sound and the rate of change of the time amplitude C:

[0023]

[0024] Where: n e is the sound end time, n s is the start time of the sound, sampling frequency F S , N is the number of sample points of signal x[n], x[n] is the nth sample value in the discrete time signal sequence, x[n+1] is the n+1th sample value in the discrete time signal sequence, and the first-order difference of signal x[n];

[0025] S202. A feature matrix is ​​set inside the feature set. Each row of each matrix represents an audio sample, and each column represents an audio feature. The audio features include fundamental frequency, dominant frequency, duration, and amplitude change. A label vector corresponding to the sound type is set based on the audio features. Each data in the label vector corresponds to the actual audio feature of an audio sample. The label vectors are pipe blockage, valve damage, and pipe aging.

[0026] S203, when training the sound recognition model, obtaining an audio data set as an input sample of the sound recognition model, and the input layer performs error removal and noise removal preprocessing operations on the input data;

[0027] Get the feature set as the comparison sample of the recognition layer, and use the label vector as the output result of the output layer;

[0028] When testing the sound recognition model, real-time audio data is obtained and input into the input layer. After the frequency features and time domain features are extracted through the recognition layer, the label vector is output according to the recognition results.

[0029] Furthermore, the specific process of performing abnormal monitoring judgment to generate abnormal monitoring signals is as follows:

[0030] S300: Compare the pressure value and sound of the smart fire hydrant obtained, compare the pressure value with the warning threshold, and compare the sound with the type of abnormal sound.

[0031] S301, preliminarily determine the pressure value y, if y < g, it is determined that the pressure value is abnormal, and a pressure value abnormality monitoring signal is generated;

[0032] S303: If the label vector output by the sound recognition model is an abnormal sound type when the acquired sound is being identified as an abnormal sound type, an abnormal monitoring signal of the corresponding type is generated.

[0033] Furthermore, a pressure value warning level is established based on the warning threshold, and a warning level report is generated. The specific process is as follows:

[0034] S401. Obtain an early warning threshold g, and determine the abnormality of the pressure value obtained in real time based on the early warning threshold, determine the development trend of the abnormality, and analyze whether the pressure change is a temporary fluctuation or a continuous deterioration;

[0035] S402, the process of predicting and calculating the trend of the pressure value is as follows: the time points t1, t2, ..., tn at which the smart fire hydrant generates the pressure value; the corresponding pressure values ​​y1, y2, ..., yn;

[0036] Preliminary calculation of the deviation d between the abnormal pressure value and the warning threshold i :y i -g;

[0037] Further calculate the change rate r of the deviation between the abnormal time point and the adjacent time point i :

[0038]

[0039] Get the average rate of change

[0040]

[0041] Get the standard deviation w of the rate of change:

[0042]

[0043] Set the standard change rate to α. If α>w, it means it is an accidental situation and is at the warning level. This means that the pressure value of the smart fire hydrant will continue to rise, reaching the alarm level. This means that the pressure value of the smart fire hydrant will continue to drop to the alarm level.

[0044] Furthermore, the specific process of obtaining the monitoring and early warning signal of abnormal intelligent fire hydrants is as follows:

[0045] Obtain the warning level and sound category report of the real-time monitored smart fire hydrant, and generate a monitoring warning signal based on the correspondence between the two;

[0046] If the warning level of the smart fire hydrant is the early warning level, smart regulation will be generated;

[0047] If the warning level of the smart fire hydrant is the alarm level, a manual control signal is generated;

[0048] If the sound of the smart fire hydrant is an abnormal sound type such as a blocked pipe or damaged valve, an inspection and maintenance signal will be generated.

[0049] Furthermore, the monitoring requirements for smart fire hydrants are as follows:

[0050] Obtain monitoring and early warning signals of abnormal smart fire hydrants. When the monitoring and early warning signal is an intelligent control signal, adopt an automatic remote control procedure. When the monitoring and early warning signal is a manual control signal, staff are required to perform manual control procedures and re-check the status of the smart fire hydrant. When an inspection and maintenance signal is generated, manual maintenance of the smart fire hydrant is required.

[0051] The multi-point pressure monitoring and early warning system for intelligent fire hydrants includes a multi-point monitoring module, an early warning analysis module, a monitoring identification module, a monitoring early warning module, and an early warning transmission module.

[0052] The multi-point monitoring module obtains the pressure value of each smart fire hydrant through a pressure sensor. A microphone sensor is installed at the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants.

[0053] The early warning analysis module is used to obtain historical data generated by smart fire hydrants from the online resource database. The historical data includes the pressure value data and sound data of the smart fire hydrants, and generates an early warning mechanism for monitoring the smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets the warning threshold based on historical data. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging.

[0054] The monitoring and identification module is used to monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and perform abnormal monitoring judgment to generate abnormal monitoring signals;

[0055] The monitoring and early warning module establishes a pressure value early warning level according to the early warning threshold, generates an early warning level report, obtains the generated abnormal monitoring signal, identifies the pressure value of the abnormal smart fire hydrant, and performs sound abnormality recognition, generates a sound abnormality type report, and generates a monitoring early warning signal of the abnormal smart fire hydrant according to the early warning level report and the abnormality type report;

[0056] The early warning transmission module is used to obtain the monitoring early warning signal of abnormal smart fire hydrants, and output it to the department management end for intelligent personnel scheduling based on the monitoring requirements of the smart fire hydrants.

[0057] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0058] The multi-point pressure monitoring and early warning method and system for smart fire hydrants monitors the pressure conditions of smart fire hydrants by obtaining the specific locations of smart fire hydrants within the farm area, sets early warning thresholds and abnormal sound types based on historical data, and predicts and determines whether the monitored real-time pressure values ​​are at abnormal pressure values ​​to generate abnormal monitoring signals. The abnormal pressure values ​​are then analyzed for early warning levels to distinguish whether the pressure drop is a temporary fluctuation or continued deterioration, and the changing trends of smart fire hydrants are effectively predicted to generate monitoring and early warning signals for abnormal smart fire hydrants. Effective resource scheduling is then carried out based on the monitoring needs of the department. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Shown is a schematic flow chart of the method of the present invention;

[0060] Figure 2 It shows a schematic diagram of the overall system distribution structure of the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] Example 1:

[0063] like Figure 1 As shown, the multi-point pressure monitoring and early warning method and system for intelligent fire hydrants include the following steps:

[0064] Step 1: When performing multi-point pressure monitoring on smart fire hydrants, the pressure value of each smart fire hydrant is obtained through a pressure sensor. A microphone sensor is installed on the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants.

[0065] Since fire hydrant pipes may have a humid environment, a waterproof microphone sensor is selected;

[0066] Step 2: Obtain historical data generated by smart fire hydrants from an online resource database. The historical data includes pressure and sound data of smart fire hydrants, and generate an early warning mechanism for monitoring smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets an early warning threshold based on historical data and uses the early warning threshold to determine abnormalities in real-time pressure values. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging.

[0067] Step 3: Monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and conduct abnormal monitoring judgment to generate abnormal monitoring signals;

[0068] Step 4: Establish a pressure value warning level based on the warning threshold, generate a warning level report, obtain the generated abnormal monitoring signal, and perform sound abnormality identification based on the pressure value of the abnormal smart fire hydrant to generate a sound abnormality type report. Based on the warning level report and the abnormality type report, obtain the monitoring warning signal of the abnormal smart fire hydrant;

[0069] Step 5: Obtain the monitoring and early warning signal of the abnormal smart fire hydrant, and output it to the department management end for intelligent personnel dispatch based on the monitoring demand personnel in the smart fire hydrant.

[0070] When obtaining the pressure value and sound of the smart fire hydrant in step 1, the obtained pressure value and sound are sent as data to the data storage control end, and the database storage control end stores and records them, and establishes a storage analysis table for each smart fire hydrant to record the pressure value and sound generated by the smart fire hydrant.

[0071] The specific process of judging abnormal conditions of real-time pressure values ​​through early warning thresholds is as follows:

[0072] Obtain the historical pressure data of the smart fire hydrants within the farm area, select the pressure value data within a period of time based on the historical pressure data, average the pressure data values ​​and remove the error, and obtain the calculated pressure value p: q is the number of pressure values ​​calculated, m = 1, 2, ..., q, p i is the pressure value of the i-th;

[0073] The specific steps for identifying abnormal sound in step 2 include:

[0074] S200: Acquire historical sound data generated by the smart fire hydrant in the event of pipe blockage, valve damage, or pipe aging, and integrate the historical sound data as audio data to obtain an audio dataset;

[0075] A sound recognition model is built based on the audio dataset using a convolutional neural network. The sound recognition model consists of an input layer, a recognition layer, and an output layer. The audio dataset is used as input. The input layer classifies and labels the audio dataset, and performs error and noise removal preprocessing on the audio data to obtain the processed audio dataset.

[0076] S201: The recognition layer obtains the processed audio data set and extracts the frequency and time domain features of the audio data to obtain the extracted feature set. The frequency features are the basic frequency and main frequency of the generated sound, and the time domain features are the duration of the sound and the time amplitude change rate. The amplitude change rate is calculated by the time change of the audio data. The specific recognition process is as follows:

[0077] Setting related functions

[0078] According to the sampling frequency F S Calculate the main frequency F:

[0079] Where: R[T] is the autocorrelation function value of signal x[n], which describes the correlation of the signal itself when the delay is K, K is the delay amount, which is used to represent the comparison of the signal itself at different time offsets, n is the index of the discrete time series, N is the number of sample points of signal x[n], K0 is the delay corresponding to the first peak in the autocorrelation function except K=0, m is the frequency domain index, X[m] is the discrete Fourier transform result, which is the mth coefficient in the frequency domain, j is a virtual unit, and M is the index corresponding to the main frequency;

[0080] Calculate the duration T of the sound and the rate of change of the time amplitude C:

[0081]

[0082] Where: n e is the sound end time, n s is the start time of the sound, sampling frequency F S , N is the number of sample points of signal x[n], x[n] is the nth sample value in the discrete time signal sequence, x[n+1] is the n+1th sample value in the discrete time signal sequence, and the first-order difference of signal x[n];

[0083] S202. A feature matrix is ​​set inside the feature set. Each row of each matrix represents an audio sample, and each column represents an audio feature. The audio features include fundamental frequency, dominant frequency, duration, and amplitude change. A label vector corresponding to the sound type is set based on the audio features. Each data in the label vector corresponds to the actual audio feature of an audio sample. The label vectors are pipe blockage, valve damage, and pipe aging.

[0084] S203, when training the sound recognition model, obtaining an audio data set as an input sample of the sound recognition model, and the input layer performs error removal and noise removal preprocessing operations on the input data;

[0085] Get the feature set as the comparison sample of the recognition layer, and use the label vector as the output result of the output layer;

[0086] When testing the sound recognition model, real-time audio data is obtained and input into the input layer. After the frequency features and time domain features are extracted through the recognition layer, the label vector is output according to the recognition results.

[0087] The specific process of performing abnormal monitoring judgment to generate abnormal monitoring signals is as follows:

[0088] S300: Compare the pressure value and sound of the smart fire hydrant obtained, compare the pressure value with the warning threshold, and compare the sound with the type of abnormal sound.

[0089] S301, preliminarily determine the pressure value y, if y < g, it is determined that the pressure value is abnormal, and a pressure value abnormality monitoring signal is generated;

[0090] S303: If the label vector output by the sound recognition model is an abnormal sound type when the acquired sound is being identified as an abnormal sound type, an abnormal monitoring signal of the corresponding type is generated.

[0091] Establish a pressure value warning level based on the warning threshold and generate a warning level report. The specific process is as follows:

[0092] Obtain the warning threshold g, and use it to determine if the real-time pressure value is abnormal, determine the development trend of the abnormality, and analyze whether the pressure change is a temporary fluctuation or a continuous deterioration;

[0093] S102, the process of predicting and calculating the trend of the pressure value is as follows: the time points t1, t2, ..., tn at which the smart fire hydrant generates the pressure value; the corresponding pressure values ​​y1, y2, ..., yn;

[0094] Preliminary calculation of the deviation d between the abnormal pressure value and the warning threshold i :y i -g;

[0095] Further calculate the change rate r of the deviation between the abnormal time point and the adjacent time point i :

[0096]

[0097] Get the average rate of change

[0098]

[0099] Get the standard deviation w of the rate of change:

[0100]

[0101] Set the standard change rate to α. If α>w, it means it is an accidental condition. This means that the pressure value of the smart fire hydrant will continue to rise. This means that the pressure value of the smart fire hydrant will continue to drop. The specific process of obtaining the monitoring and early warning signal of abnormal smart fire hydrants is as follows:

[0102] S402: Obtain a warning level and sound category report of the real-time monitored smart fire hydrant, and generate a monitoring warning signal based on the correspondence between the two.

[0103] If the warning level of the smart fire hydrant is the early warning level, smart regulation will be generated;

[0104] If the warning level of the smart fire hydrant is the alarm level, a manual control signal is generated;

[0105] If the warning level of the smart fire hydrant is a fault level and the sound is an abnormal sound type such as a pipe blockage or a damaged valve, an inspection and maintenance signal will be generated.

[0106] The specific personnel who need monitoring in smart fire hydrants are as follows:

[0107] Obtain monitoring and early warning signals of abnormal smart fire hydrants. When the monitoring and early warning signal is an intelligent control signal, adopt an automatic remote control procedure. When the monitoring and early warning signal is a manual control signal, staff are required to perform manual control procedures and re-check the status of the smart fire hydrant. When an inspection and maintenance signal is generated, manual maintenance of the smart fire hydrant is required.

[0108] Example 2:

[0109] like Figure 2 As shown, the multi-point pressure monitoring and early warning system for intelligent fire hydrants includes a multi-point monitoring module, an early warning analysis module, a monitoring identification module, a monitoring early warning module, and an early warning transmission module.

[0110] The multi-point monitoring module obtains the pressure value of each smart fire hydrant through a pressure sensor. A microphone sensor is installed at the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants.

[0111] The early warning analysis module is used to obtain historical data generated by smart fire hydrants from the online resource database. The historical data includes the pressure value data and sound data of the smart fire hydrants, and generates an early warning mechanism for monitoring the smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets the warning threshold based on historical data. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging.

[0112] The monitoring and identification module is used to monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and perform abnormal monitoring judgment to generate abnormal monitoring signals;

[0113] The monitoring and early warning module establishes a pressure value early warning level according to the early warning threshold, generates an early warning level report, obtains the generated abnormal monitoring signal, identifies the pressure value of the abnormal smart fire hydrant, and performs sound abnormality recognition, generates a sound abnormality type report, and generates a monitoring early warning signal of the abnormal smart fire hydrant according to the early warning level report and the abnormality type report;

[0114] The early warning transmission module is used to obtain the monitoring early warning signal of abnormal smart fire hydrants, and output it to the department management end for intelligent personnel dispatch according to the monitoring demand personnel in the smart fire hydrants;

[0115] The department management end is responsible for managing the personnel who monitor smart fire hydrants, issuing tasks, and scheduling work tasks.

[0116] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0117] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by those skilled in the art according to actual conditions.

[0118] In the two embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the method embodiments described above are merely illustrative. For example, the module division is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not performed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, method or module, and may be electrical, mechanical or other forms.

[0119] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A multi-point pressure monitoring and early warning method for smart fire hydrants, characterized in that: The following steps are involved: Step 1: When performing multi-point pressure monitoring on smart fire hydrants, the pressure value of each smart fire hydrant is obtained through a pressure sensor. A microphone sensor is installed on the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants. Step 2: Obtain historical data generated by smart fire hydrants from an online resource database. The historical data includes pressure and sound data of smart fire hydrants, and generate an early warning mechanism for monitoring smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets an early warning threshold based on historical data and uses the early warning threshold to determine abnormalities in real-time pressure values. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging. Step 3: Monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and conduct abnormal monitoring judgment to generate abnormal monitoring signals; Step 4: Establish a pressure value warning level based on the warning threshold, generate a warning level report, obtain the generated abnormal monitoring signal, and perform sound abnormality identification based on the pressure value of the abnormal smart fire hydrant to generate a sound abnormality type report. Based on the warning level report and the abnormality type report, obtain the monitoring warning signal of the abnormal smart fire hydrant; Step 5: Obtain the monitoring and early warning signal of the abnormal smart fire hydrant, and output it to the department management end for intelligent personnel dispatch based on the monitoring demand personnel in the smart fire hydrant.

2. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: When obtaining the pressure value and sound of the smart fire hydrant in step 1, the obtained pressure value and sound are sent as data to the data storage control end, and the database storage control end stores and records them, and establishes a storage analysis table for each smart fire hydrant to record the pressure value and sound generated by the smart fire hydrant.

3. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: The specific process of judging abnormal conditions of real-time pressure values ​​through early warning thresholds is as follows: Obtain the historical pressure data of the smart fire hydrants within the farm area, select the pressure value data within a period of time based on the historical pressure data, average the pressure data values ​​and remove the error, and obtain the calculated pressure value p: q is the number of pressure values ​​calculated, m = 1, 2, ..., q, p i is the pressure value of the ith one.

4. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: The specific steps for identifying abnormal sound in step 2 include: S200: Acquire historical sound data generated by the smart fire hydrant in the event of pipe blockage, valve damage, or pipe aging, and integrate the historical sound data as audio data to obtain an audio dataset; A sound recognition model is built based on the audio dataset using a convolutional neural network. The sound recognition model consists of an input layer, a recognition layer, and an output layer. The audio dataset is used as input. The input layer classifies and labels the audio dataset, and performs error and noise removal preprocessing on the audio data to obtain the processed audio dataset. S201: The recognition layer obtains the processed audio data set and extracts the frequency and time domain features of the audio data to obtain the extracted feature set. The frequency features are the basic frequency and main frequency of the generated sound, and the time domain features are the duration of the sound and the time amplitude change rate. The amplitude change rate is calculated by the time change of the audio data. The specific recognition process is as follows: Set the correlation function R[K]: <Get the fundamental frequency f: According to the sampling frequency F S Calculate the main frequency F: Where: R[T] is the autocorrelation function value of signal x[n], which describes the correlation of the signal itself when the delay is K, K is the delay amount, which is used to represent the comparison of the signal itself at different time offsets, n is the index of the discrete time series, N is the number of sample points of signal x[n], K0 is the delay corresponding to the first peak in the autocorrelation function except K=0, m is the frequency domain index, X[m] is the discrete Fourier transform result, which is the mth coefficient in the frequency domain, j is a virtual unit, and M is the index corresponding to the main frequency; Calculate the duration T of the sound and the rate of change of the time amplitude C: Where: n e is the sound end time, n s is the start time of the sound, sampling frequency F S , N is the number of sample points of signal x[n], x[n] is the nth sample value in the discrete time signal sequence, x[n+1] is the n+1th sample value in the discrete time signal sequence, and the first-order difference of signal x[n]; S202. A feature matrix is ​​set inside the feature set. Each row of each matrix represents an audio sample, and each column represents an audio feature. The audio features include fundamental frequency, dominant frequency, duration, and amplitude change. A label vector corresponding to the sound type is set based on the audio features. Each data in the label vector corresponds to the actual audio feature of an audio sample. The label vectors are pipe blockage, valve damage, and pipe aging. S203, when training the sound recognition model, obtaining an audio data set as an input sample of the sound recognition model, and the input layer performs error removal and noise removal preprocessing operations on the input data; Get the feature set as the comparison sample of the recognition layer, and use the label vector as the output result of the output layer; When testing the sound recognition model, real-time audio data is obtained and input into the input layer. After the frequency features and time domain features are extracted through the recognition layer, the label vector is output according to the recognition results.

5. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: The specific process of performing abnormal monitoring judgment to generate abnormal monitoring signals is as follows: S300: Compare the pressure value and sound of the smart fire hydrant obtained, compare the pressure value with the warning threshold, and compare the sound with the type of abnormal sound. S301, preliminarily determine the pressure value y, if y < g, it is determined that the pressure value is abnormal, and a pressure value abnormality monitoring signal is generated; S303: If the label vector output by the sound recognition model is an abnormal sound type when the acquired sound is being identified as an abnormal sound type, an abnormal monitoring signal of the corresponding type is generated.

6. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: Establish a pressure value warning level based on the warning threshold and generate a warning level report. The specific process is as follows: S401. Obtain an early warning threshold g, and determine the abnormality of the pressure value obtained in real time based on the early warning threshold, determine the development trend of the abnormality, and analyze whether the pressure change is a temporary fluctuation or a continuous deterioration; S402, the process of predicting and calculating the trend of the pressure value is as follows: the time points t1, t2, ..., tn at which the smart fire hydrant generates the pressure value; the corresponding pressure values ​​y1, y2, ..., yn; Preliminary calculation of the deviation d between the abnormal pressure value and the warning threshold i :y i -g; Further calculate the change rate r of the deviation between the abnormal time point and the adjacent time point i : Get the average rate of change Get the standard deviation w of the rate of change: Set the standard change rate to α. If α>w, it means it is an accidental situation and is at the warning level. This means that the pressure value of the smart fire hydrant will continue to rise, reaching the alarm level. This means that the pressure value of the smart fire hydrant will continue to drop to the alarm level.

7. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 1 is characterized in that: The specific process of obtaining the monitoring and early warning signal of abnormal intelligent fire hydrants is as follows: Obtain the warning level and sound category report of the real-time monitored smart fire hydrant, and generate a monitoring warning signal based on the correspondence between the two; If the warning level of the smart fire hydrant is the early warning level, smart regulation will be generated; If the warning level of the smart fire hydrant is the alarm level, a manual control signal is generated; If the sound of the smart fire hydrant is an abnormal sound type such as a blocked pipe or damaged valve, an inspection and maintenance signal will be generated.

8. The multi-point pressure monitoring and early warning method for smart fire hydrants according to claim 7 is characterized in that: The specific personnel who need monitoring in smart fire hydrants are as follows: Obtain monitoring and early warning signals of abnormal smart fire hydrants. When the monitoring and early warning signal is an intelligent control signal, adopt an automatic remote control procedure. When the monitoring and early warning signal is a manual control signal, staff are required to perform manual control procedures and re-check the status of the smart fire hydrant. When an inspection and maintenance signal is generated, manual maintenance of the smart fire hydrant is required.

9. Multi-point pressure monitoring and early warning system for intelligent fire hydrants, characterized in that: Including multi-point monitoring module, early warning analysis module, monitoring identification module, monitoring early warning module, early warning transmission module, The multi-point monitoring module obtains the pressure value of each smart fire hydrant through a pressure sensor. A microphone sensor is installed at the smart fire hydrant pipe to obtain sound. The farm area is obtained to delineate the supervision area of ​​the smart fire hydrant, and a three-dimensional regional monitoring map is drawn. Different spatial coordinates and monitoring points are assigned within the three-dimensional regional monitoring map according to the different locations of the smart fire hydrants. The early warning analysis module is used to obtain historical data generated by smart fire hydrants from the online resource database. The historical data includes the pressure value data and sound data of the smart fire hydrants, and generates an early warning mechanism for monitoring the smart fire hydrants. The early warning mechanism includes pressure anomaly recognition and sound anomaly recognition. Pressure anomaly recognition sets the warning threshold based on historical data. Sound anomaly recognition includes three types of abnormal sounds: pipe blockage, valve damage, and pipe aging. The monitoring and identification module is used to monitor the smart fire hydrants in the supervision area, monitor the real-time pressure and sound of the smart fire hydrants according to the early warning mechanism, identify the pressure and sound of the smart fire hydrants, and perform abnormal monitoring judgment to generate abnormal monitoring signals; The monitoring and early warning module establishes a pressure value early warning level according to the early warning threshold, generates an early warning level report, obtains the generated abnormal monitoring signal, identifies the pressure value of the abnormal smart fire hydrant, and performs sound abnormality recognition, generates a sound abnormality type report, and generates a monitoring early warning signal of the abnormal smart fire hydrant according to the early warning level report and the abnormality type report; The early warning transmission module is used to obtain the monitoring early warning signal of abnormal smart fire hydrants, and output it to the department management end for intelligent personnel scheduling based on the monitoring requirements of the smart fire hydrants.

Images