Rain-sprinkling alarm valve group operation state remote monitoring method and system

Abstract

The application provides a rain alarm valve group operation state remote monitoring method and system. The method comprises the following steps: determining the spatial coordinates and the global diffusion direction vector of the target helium concentration abnormal diffusion area inside the sealed cavity based on the helium concentration distribution data; constructing a target composite data frame based on the phase characteristics and the helium concentration distribution data; determining the global leakage rate gradient of the target composite data frame in combination with the spatial coordinates and the global diffusion direction vector; when the global leakage rate gradient exceeds the preset reference curve for multiple times in succession, extracting the multiple harmonic distortion amplitude increment; determining the leakage level identifier according to the multiple harmonic distortion amplitude increment and the global leakage rate gradient; generating a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector; and triggering the operation state closed-loop control process of the rain alarm valve group through the instruction. The technical scheme provided by the application generates a linkage instruction after determining the leakage level, triggers the closed-loop control of the rain alarm valve group, and realizes accurate leakage response.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire alarm, and in particular to a rain alarm valve group operation state remote monitoring method and system. BACKGROUND

[0002] In an automatic sprinkler system, the rain alarm valve group as a key component, its sealing performance is directly related to the reliability and safety of the system. Due to the rain alarm valve group is in a high pressure, high humidity working environment for a long time, the helium concentration distribution in the sealed cavity, water pressure fluctuation and the change of current phase characteristics of the fire pump station and other parameters may indicate potential leakage risk. Therefore, real-time monitoring of these parameters and accurately judging the leakage situation become the key technical requirements to ensure the normal operation of fire facilities. Especially in large industrial facilities or high-rise buildings, the leakage of the rain alarm valve group may cause serious consequences, so an efficient and accurate monitoring and early warning mechanism is needed.

[0003] The existing scheme arranges helium concentration sensors, water pressure sensors inside the sealed cavity and installs current phase monitoring devices at the fire pump station to collect relevant data in real time. The collected helium concentration distribution, water pressure fluctuation and current phase characteristics are analyzed by using machine learning algorithm, a data model is constructed, and whether there is a leakage risk is judged by a pre-set threshold. The scheme can realize automatic monitoring to a certain extent and reduce the need for manual intervention.

[0004] The existing scheme realizes automatic monitoring through multi-sensor data fusion and machine learning algorithm, but it still has some shortcomings. The scheme has high real-time requirements for data processing, but in actual application, due to the large amount of data and high computational complexity, it is often difficult to realize efficient real-time analysis, resulting in delay of leakage warning. Secondly, the existing scheme relies on pre-set static threshold for judgment, which cannot dynamically adapt to complex and variable working conditions, and is easy to produce false alarm or miss alarm. The scheme lacks accurate prediction ability of leakage diffusion direction and rate, and it is difficult to provide comprehensive leakage level evaluation and accurate linkage control instructions, which limits its actual application effect in complex automatic sprinkler systems. SUMMARY

[0005] The rain alarm valve group operation state remote monitoring method and system provided by the embodiments of the present application solve the problem that the existing technology has poor accurate prediction ability of leakage diffusion direction and rate, and it is difficult to provide comprehensive leakage level evaluation and accurate linkage control instructions.

[0006] In a first aspect, the embodiments of the present application provide a rain alarm valve group operation state remote monitoring method, comprising:

[0007] Obtain helium concentration distribution data inside the sealed cavity of the deluge alarm valve group, water pressure fluctuation parameters inside the deluge alarm valve group, and current phase characteristics of the fire pump station, which is a water supply power source and establishes a protocol cascade with the deluge alarm valve group;

[0008] Determine the spatial coordinates and global diffusion direction vector of the target helium concentration abnormal diffusion area inside the sealed cavity based on the helium concentration distribution data, and construct a target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data.

[0009] Determine the global leakage rate gradient of the target composite data frame in combination with the spatial coordinates and global diffusion direction vector, extract the multiple harmonic distortion amplification quantities in the current phase characteristics when the global leakage rate gradient exceeds the preset reference curve for multiple times, and determine the leakage level identifier based on the multiple harmonic distortion amplification quantities and the global leakage rate gradient.

[0010] Generate a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector, and trigger the running state closed-loop control process of the deluge alarm valve group through the linkage instruction.

[0011] Optionally, the determination of the spatial coordinates and global diffusion direction vector of the target helium concentration abnormal diffusion area inside the sealed cavity based on the helium concentration distribution data, and the construction of the target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data, comprises:

[0012] Perform noise suppression processing on the helium concentration distribution data to generate denoised helium concentration distribution data, perform three-dimensional spatial gridding on the denoised helium concentration distribution data according to the geometric structure of the sealed cavity to generate uniformly distributed grid cells, calculate the helium concentration gradient value of each grid cell, and identify the initial helium concentration abnormal diffusion area.

[0013] Perform region growing processing on the initial helium concentration abnormal diffusion area, and during the region growing processing, merge adjacent grid cells based on a preset gradient direction consistency criterion until the gradient value at the growth boundary is lower than a preset gradient threshold value to obtain the target helium concentration abnormal diffusion area and determine the spatial coordinates of the target helium concentration abnormal diffusion area.

[0014] Select a plurality of grid cells with larger gradient values in the target helium concentration abnormal diffusion area as a plurality of representative grid cells, calculate the diffusion direction vector of each representative grid cell, and perform normalized weighted average processing on all the diffusion direction vectors to generate the global diffusion direction vector of the target helium concentration abnormal diffusion area.

[0015] performing Fourier transform on the water pressure fluctuation parameter, extracting a phase feature in a specific frequency range matched with a resonance frequency of the sealed cavity structure from a Fourier transform result, performing sliding window smoothing processing and low-pass filtering operation on the phase feature in the specific frequency range, and generating a denoised phase feature as a phase feature of the water pressure fluctuation parameter;

[0016] calculating water pressure phase fluctuation amplitudes of each of the grid units in a corresponding time window according to the phase feature of the water pressure fluctuation parameter;

[0017] constructing a three-dimensional coupling coefficient matrix based on the helium concentration gradient values of each of the grid units and the water pressure phase fluctuation amplitudes, and performing tensor product operation on the three-dimensional coupling coefficient matrix and the global diffusion direction vector to generate a target composite data frame.

[0018] Optionally, the constructing a three-dimensional coupling coefficient matrix based on the helium concentration gradient values of each of the grid units and the water pressure phase fluctuation amplitudes, and performing tensor product operation on the three-dimensional coupling coefficient matrix and the global diffusion direction vector to generate a target composite data frame comprises:

[0019] performing normalization processing on the helium concentration gradient values of each of the grid units respectively to generate a normalized helium concentration gradient value sequence, and performing standardization processing on the water pressure phase fluctuation amplitudes of each of the grid units in a corresponding time window to generate a normalized water pressure phase fluctuation amplitude sequence;

[0020] calculating a gradient phase coupling coefficient of each grid unit based on the helium concentration gradient value sequence and the water pressure phase fluctuation amplitude sequence, the gradient phase coupling coefficient being obtained through multiplication operation of the helium concentration gradient value and the water pressure phase fluctuation amplitude;

[0021] arranging the gradient phase coupling coefficients of all the grid units according to spatial positions to construct a three-dimensional coupling coefficient matrix, and performing tensor extension on the global diffusion direction vector to generate a diffusion direction tensor;

[0022] performing element-by-element multiplication operation on the three-dimensional coupling coefficient matrix and the diffusion direction tensor to generate an initial composite data frame;

[0023] performing spatial smoothing processing on the initial composite data frame to eliminate local data mutations to generate a target composite data frame.

[0024] Optionally, the performing element-by-element multiplication operation on the three-dimensional coupling coefficient matrix and the diffusion direction tensor to generate an initial composite data frame comprises:

[0025] dimensionally check the three-dimensional coupling coefficient matrix to ensure that the three-dimensional coupling coefficient matrix is consistent with the spatial dimension of the diffusion direction tensor;

[0026] weight correct each element in the three-dimensional coupling coefficient matrix to adjust the weight value of the gradient phase coupling coefficient, generate a corrected three-dimensional coupling coefficient matrix, normalize the diffusion direction tensor to ensure that the modulus of the diffusion direction vector of each representative grid cell is a unit length, and generate a normalized diffusion direction tensor;

[0027] perform element-by-element multiplication operation on the corrected three-dimensional coupling coefficient matrix and the normalized diffusion direction tensor to generate an initial composite data frame.

[0028] Optionally, the initial composite data frame is spatially smoothed to eliminate local data mutations and generate a target composite data frame, including:

[0029] perform local neighborhood analysis on the helium concentration gradient value of each grid cell in the initial composite data frame, and in the local neighborhood analysis process, calculate the difference in helium concentration gradient value between each grid cell and its adjacent grid cells to generate a local data difference matrix;

[0030] Based on the local data difference matrix, identify grid cells with a helium concentration gradient value difference exceeding a preset difference threshold, mark them as local mutation grid cells, and perform interpolation correction on the helium concentration gradient value of the local mutation grid cells to generate a corrected helium concentration gradient value;

[0031] update the corrected helium concentration gradient value to the initial composite data frame to generate an intermediate composite data frame, perform smoothing on the intermediate composite data frame, adjust the helium concentration gradient value of each grid cell based on a preset spatial continuity constraint, and generate a smoothed composite data frame;

[0032] Perform boundary consistency check on the smoothed composite data frame, correct abnormal data points at the boundary of each local mutation grid cell based on the gradient change of the data value of adjacent grid cells, and generate a target composite data frame.

[0033] Optionally, the spatial coordinates and global diffusion direction vector are combined to determine the global leakage rate gradient of the target composite data frame, and when the global leakage rate gradient exceeds a preset reference curve for multiple times in succession, the multiple harmonic distortion amplification amount in the current phase feature is extracted, and the leakage level identifier is determined according to the multiple harmonic distortion amplification amount and the global leakage rate gradient, including:

[0034] determine a global leakage rate gradient of the target composite data frame based on the time sequence of the leakage rate gradient of each grid cell and in combination with a global diffusion direction vector;

[0035] determine whether the global leakage rate gradient exceeds a preset reference curve for multiple times in succession;

[0036] if yes, determine an abnormal time window, the abnormal time window being a time window corresponding to a time period during which the global leakage rate gradient exceeds the preset reference curve for multiple times in succession;

[0037] extract a multiple harmonic distortion amplification amount in the current phase feature within the abnormal time window, and generate a harmonic distortion accumulation amount by time integration of the multiple harmonic distortion amplification amount;

[0038] determine a leakage level identifier based on a ratio relationship between the harmonic distortion accumulation amount and the global leakage rate gradient.

[0039] Optionally, the determining of the global leakage rate gradient of the target composite data frame based on the time sequence of the leakage rate gradient and in combination with a global diffusion direction vector comprises:

[0040] calculate a helium concentration gradient value change rate between grid cells according to the helium concentration gradient value of each grid cell;

[0041] calculate a weight of each grid cell according to the helium concentration gradient value change rate, in combination with a global diffusion direction vector, a distance from the grid cell to a leakage source, and an estimated diffusion speed of helium in the target helium concentration abnormal diffusion area;

[0042] optimize the weight of each grid cell in combination with historical data to obtain an optimized weight of each grid cell, so as to construct a weighted matrix;

[0043] combine the time sequence of the leakage rate gradient with the weighted matrix to obtain the global leakage rate gradient of the target composite data frame.

[0044] In a second aspect, the embodiments of the present application provide a rain-sprinkling alarm valve group operation state remote monitoring system,

[0045] an acquisition module: acquire helium concentration distribution data inside a sealed cavity of a rain-sprinkling alarm valve group, water pressure fluctuation parameters inside the rain-sprinkling alarm valve group, and current phase features of a fire pump station, the fire pump station being a water supply power source and establishing a protocol cascade with the rain-sprinkling alarm valve group;

[0046] The construction module is configured to determine the spatial coordinates and the global diffusion direction vector of the target helium concentration abnormal diffusion area inside the sealed cavity based on the helium concentration distribution data, and construct a target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data.

[0047] The extraction module is configured to determine the global leakage rate gradient of the target composite data frame in combination with the spatial coordinates and the global diffusion direction vector, extract the multiple harmonic distortion amplification quantities in the current phase characteristics when the global leakage rate gradient continuously exceeds the preset reference curve for multiple times, and determine the leakage level identifier according to the multiple harmonic distortion amplification quantities and the global leakage rate gradient.

[0048] The generation module is configured to generate a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector, and trigger the operation state closed-loop control process of the rain-sprinkling alarm valve group through the linkage instruction.

[0049] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the production intelligent monitoring method based on the industrial internet according to any one of the first aspect.

[0050] In a fourth aspect, an embodiment of the present application provides a computing device and a computer storage medium storing a computer program, when the computer program is executed by a computer, a production intelligent monitoring method based on the industrial internet according to any one of the first aspect is implemented.

[0051] In the embodiment of the present application, a rain-sprinkling alarm valve group operation state remote monitoring method is provided, and the method includes: acquiring helium concentration distribution data inside a sealed cavity of a rain-sprinkling alarm valve group, water pressure fluctuation parameters inside the rain-sprinkling alarm valve group, and current phase characteristics of a fire pump station; the fire pump station is used as a water supply power source and is connected with the rain-sprinkling alarm valve group in a protocol linkage; spatial coordinates and a global diffusion direction vector of a target helium concentration abnormal diffusion area inside the sealed cavity are determined based on the helium concentration distribution data; a target composite data frame is constructed based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data; a global leakage rate gradient of the target composite data frame is determined in combination with the spatial coordinates and the global diffusion direction vector; when the global leakage rate gradient continuously exceeds a preset reference curve for multiple times, multiple harmonic distortion amplification quantities in the current phase characteristics are extracted; a leakage level identifier is determined according to the multiple harmonic distortion amplification quantities and the global leakage rate gradient; a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector is generated; and an operation state closed-loop control process of the rain-sprinkling alarm valve group is triggered through the linkage instruction.

[0052] The rain alarm valve group operation state remote monitoring method provided by the embodiments of the present application can not only monitor potential leakage risks, but also accurately determine the specific location and diffusion direction of the leakage based on the comprehensive data. Compared with the leakage judgment method relying on static threshold in the prior art, the method improves the accuracy and real-time performance of detection. The embodiments of the present application use the multiple harmonic distortion amplitude increment and the global leakage rate gradient to determine the leakage level identifier, thereby realizing dynamic evaluation of the leakage condition. This enables the embodiments of the present application to better adapt to different working conditions and reduce false positives and false negatives. The present application can not only accurately locate the leakage point and predict the direction and speed of leakage diffusion, but also generate a linkage instruction containing the leakage level identifier, spatial coordinates and global diffusion direction vector according to the information, and trigger the operation state closed-loop control process of the rain alarm valve group. This method provides more comprehensive leakage evaluation and can more accurately execute the corresponding control measures, effectively improving the response efficiency and reliability of the automatic sprinkler fire extinguishing system. Further, the embodiments of the present application perform noise suppression and three-dimensional gridding processing on the helium concentration distribution data, identify the helium concentration abnormal diffusion area and calculate its spatial coordinates and diffusion direction vector; perform Fourier transform and filtering processing on the water pressure fluctuation parameters to extract the phase characteristics, and construct a three-dimensional coupling coefficient matrix combined with the helium concentration gradient, which improves the accuracy and real-time performance of the rain alarm valve group leakage monitoring, and is suitable for the demand of the automatic sprinkler fire extinguishing system under complex working conditions.

[0053] These aspects or other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0055] Figure 1 A flow chart of a valve group operation state remote monitoring method provided by the embodiments of the present application;

[0056] Figure 2 A structure schematic diagram of a valve group operation state remote monitoring system provided by the embodiments of the present application;

[0057] Figure 3 A structure schematic diagram of a computing device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0058] In order for those skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0059] In some processes described in the specification and claims of the present application and the above description, a plurality of operations appearing in a specific order are included, but it should be clearly understood that these operations can be executed in the order appearing in the text or in parallel, and the serial numbers of the operations such as 11, 12, etc. are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in the text are used to distinguish different messages, devices, modules, etc. and do not represent the order and do not limit that "first" and "second" are different types.

[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0061] Figure 1 A flow chart of a valve group operation state remote monitoring method provided by the embodiments of the present application is shown in FIG. 1, which comprises the following steps. Figure 1

[0062] S11, obtain the helium concentration distribution data inside the sealed cavity of the deluge alarm valve group, the water pressure fluctuation parameters inside the deluge alarm valve group, and the current phase characteristics of the fire pump station. The fire pump station serves as a water supply power source and establishes a protocol cascade with the deluge alarm valve group.

[0063] Among them, the helium concentration distribution data is obtained by a helium concentration sensor arranged in the sealed cavity, which is used to reflect the distribution of helium in the cavity and serves as a key indicator for leakage detection. The water pressure fluctuation parameters are collected by a water pressure sensor, which are used to reflect the dynamic changes of the water pressure inside the deluge alarm valve group and can assist in judging the leakage condition in combination with the helium concentration data. The current phase characteristics are obtained by a current monitoring device of the fire pump station, which are used to analyze the operation state of the pump station, and the phase change can indirectly reflect the influence of leakage on the water supply system.

[0064] S12, determine the spatial coordinates of the target helium concentration abnormal diffusion area inside the sealed cavity and the global diffusion direction vector based on the helium concentration distribution data, and construct a target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data. ​

[0065] The target helium concentration abnormal diffusion area is an area with a helium concentration higher than a normal value, which is identified by the helium concentration distribution data, and the area can be a leakage source. The global diffusion direction vector reflects the diffusion direction of helium in the sealed cavity, and is used to predict the leakage trend. The target composite data frame is a data structure generated by fusing the helium concentration distribution data and the phase feature of the water pressure fluctuation parameter, and is used to comprehensively judge the leakage condition.

[0066] S13, in combination with the spatial coordinates and the global diffusion direction vector, determine the global leakage rate gradient of the target composite data frame. When the global leakage rate gradient exceeds the preset reference curve continuously for multiple times, extract the multiple harmonic distortion amplitude increase amount in the current phase feature, and determine the leakage level identifier according to the multiple harmonic distortion amplitude increase amount and the global leakage rate gradient.

[0067] The global leakage rate gradient reflects the change trend of the leakage rate in space and time, and is used to quantify the leakage severity. The multiple harmonic distortion amplitude increase amount is an abnormal change amount of the harmonic component in the current phase feature, and is used to assist in judging the influence of the leakage on the water supply system. The leakage level identifier is a leakage severity level determined according to the leakage rate gradient and the harmonic distortion amplitude increase amount. For example, the leakage level identifier is low leakage level V1, medium leakage level V2, medium-high leakage level V3, high leakage level V4, and serious leakage level V5.

[0068] S14, generate a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector, and trigger the operation state closed-loop control process of the deluge alarm valve group through the linkage instruction.

[0069] The linkage instruction includes control instructions of the leakage level, the leakage source position, and the diffusion direction, and is used to trigger the closed-loop control of the deluge alarm valve group. The association between the running state of the deluge alarm valve group and the target helium concentration abnormal diffusion area mainly reflects the leakage detection and positioning capability of the system, especially in the monitoring and maintenance stage. Specifically, for the leakage detection, the following is explained: In the automatic sprinkler system, helium is used as a tracer gas to detect small leaks in sealed cavities. By monitoring the change of helium concentration, it can be found whether there is a leakage. Due to its small molecule, inertness and easy detection, helium becomes an ideal leakage test medium. For the leakage position determination process, the following is explained: Once the abnormal increase of helium concentration is detected, it indicates that there is a leakage point (i.e. the leakage source). Based on the helium concentration distribution data, the spatial coordinates of the target helium concentration abnormal diffusion area and the diffusion direction vector can be calculated. This helps to accurately locate the specific position of the leakage. For the influence of the running state, the following is explained: In the normal standby state of the deluge alarm valve group, if the target helium concentration abnormal diffusion area is monitored, it may indicate that there is a potential leakage risk in the system, which needs to be further checked and handled to avoid affecting the reliability of the automatic sprinkler system. In the activated or water flow state of the deluge alarm valve group, if the helium concentration anomaly is detected, in addition to focusing on the fire response, it also needs to consider whether the problem is caused by internal leakage of the automatic sprinkler system, which may affect the fire extinguishing efficiency. In the fault state of the deluge alarm valve group, the helium concentration anomaly may be one of the causes of the fault, such as seal failure. In the embodiments of the present application, timely identification and repair of the target helium concentration abnormal diffusion area can help restore the normal function of the system. During the test state of the deluge alarm valve group, helium concentration monitoring can be used to verify the integrity of the automatic sprinkler system, ensuring that all components are well sealed and no leakage occurs. The closed-loop control process automatically adjusts the running state of the deluge alarm valve group according to the linkage instruction to cope with the leakage.

[0070] The following is a specific example:

[0071] In the automatic sprinkler system of a certain high-rise building, helium concentration anomaly is detected in the sealed cavity of the deluge alarm valve group. The remote monitoring system first obtains the distribution data through the helium concentration sensor, and combines the water pressure fluctuation parameters and the current phase characteristics of the fire pump station to identify that the leakage source is located at the northeast corner of the cavity. The spatial coordinates of the leakage area are determined by the region growing algorithm, and the global diffusion direction vector is calculated. Further, the remote monitoring system detects that the leakage rate gradient exceeds the baseline curve for three consecutive times, and harmonic distortion appears in the current phase characteristics, determining that the leakage level is "serious". The remote monitoring system generates linkage instructions to trigger the closed-loop control process of the deluge alarm valve group, automatically closing the leakage area valve and starting the standby pump station to ensure the normal operation of the automatic sprinkler system.

[0072] By performing S11-S14, the present embodiment achieves precise monitoring and efficient control of the leakage of the rain-sprinkler valve group through multi-source data fusion and fine processing. The specific effects include: accurately locating the leakage source, dynamically reflecting the leakage trend, avoiding false positives and false negatives, and quickly responding to leakage events through closed-loop control, thereby improving the reliability and safety of the automatic sprinkler system, and being suitable for leakage monitoring and control requirements under complex conditions.

[0073] In one possible embodiment, S12, based on the helium concentration distribution data, determines the spatial coordinates and global diffusion direction vector of the target helium concentration abnormal diffusion region inside the sealed cavity, and based on the phase characteristics of the water pressure fluctuation parameter and the helium concentration distribution data, constructs a target composite data frame, including:

[0074] Step 121, noise suppression processing is performed on the helium concentration distribution data to generate denoised helium concentration distribution data, the denoised helium concentration distribution data is three-dimensionally gridded according to the geometric structure of the sealed cavity to generate uniformly distributed grid cells, the helium concentration gradient value of each grid cell is calculated, and the initial helium concentration abnormal diffusion region is identified.

[0075] The noise suppression processing refers to removing noise interference in the helium concentration distribution data through a filtering algorithm to improve data quality. Three-dimensional spatial gridding divides the sealed cavity into uniform three-dimensional grid cells, facilitating subsequent data analysis and calculation. The helium concentration gradient value reflects the rate of change of helium concentration in space and is used to identify abnormal diffusion regions. The initial helium concentration abnormal diffusion region is a possible leakage area identified by the gradient value. The initial target helium concentration abnormal diffusion region is an abnormal area in the helium concentration distribution that can be identified by analyzing the helium concentration gradient value of each grid cell. These regions usually have a gradient value higher than that of the surrounding area, and these regions may be leakage points or diffusion sources. The diffusion source is the area around the leakage point. Each grid cell is calculated by dividing the sealed cavity into uniform three-dimensional grid cells, each cell coordinate is (i,j,k), and the corresponding helium concentration is C(i,j,k) (denoised data). Then the central difference method is used to calculate the gradient vector and the amplitude of the gradient value of each grid cell in three-dimensional space, and then the boundary processing method is used to obtain the gradient value.

[0076] Step 122, region growing processing is performed on the initial helium concentration abnormal diffusion region. During the region growing processing, adjacent grid cells are merged based on a preset gradient direction consistency criterion until the gradient value at the growth boundary is lower than a preset gradient threshold value, to obtain the target helium concentration abnormal diffusion region and determine the spatial coordinates of the target helium concentration abnormal diffusion region.

[0077] The region growing processing refers to gradually expanding the abnormal region by merging adjacent grid cells until a preset stop condition is met. The preset stop condition is that when the helium concentration gradient amplitude of adjacent grid cells at the growth boundary is lower than a preset gradient threshold, the growth is stopped. The gradient direction consistency criterion is used as a standard for judging whether adjacent grid cells belong to the same abnormal region. Adjacent grid cells in a three-dimensional grid refer to cells that are in direct contact with the current target grid cell and share at least one face, forming a direct spatial adjacency relationship. The adjacent grid cells are merged. A specific example is as follows: assuming that the initial abnormal region is a high-gradient cell A, and there is a cell B in the six-neighborhood of A. If the gradient amplitude of B is higher than the threshold, and the gradient direction of B is highly consistent with the average direction of region A (the cosine value of the included angle between the directions is greater than or equal to 0.9), then B will be merged into the abnormal region of A. After merging, the average gradient direction of the region will be updated to the average value of the gradient vectors of A and B, and the spatial boundary coordinates will also be expanded to include the position of B. Subsequent iterations will continue to grow outward with B as the seed point.

[0078] The final leakage region determined after the region growing processing of the target helium concentration abnormal diffusion region. Through the region growing processing and the gradient direction consistency criterion, the target helium concentration abnormal diffusion region can be further accurately located, and its spatial coordinates and diffusion direction can be determined.

[0079] Step 123, selecting a plurality of grid cells with large gradient values in the target helium concentration abnormal diffusion region as a plurality of representative grid cells, calculating the diffusion direction vector of each representative grid cell, and performing normalized weighted average processing on all diffusion direction vectors to generate a global diffusion direction vector of the target helium concentration abnormal diffusion region.

[0080] The representative grid cell is a grid cell with a large gradient value, which is used to reflect the main direction of the leakage diffusion. The diffusion direction vector can reflect the diffusion direction of helium in the representative grid cell. The global diffusion direction vector is a comprehensive diffusion direction generated by normalized weighted average, which is used to predict the leakage trend. Normalized weighted average refers to a method of weighted average of multiple vectors, and the core steps include vector normalization, weight distribution, and weighted average.

[0081] The following is a specific example: assume that there are 3 representative grid units in the target abnormal area, and their diffusion direction vectors and gradient values (as weights) are as follows: unit A's original direction vector (3, 0, 0), gradient value wA = 5, unit B's original direction vector (0, 4, 0), gradient value wB = 3, unit C's original direction vector (0, 0, 5), gradient value wC = 2. Vector normalization is to convert each vector into a unit vector, so the vector corresponding to unit A (3, 0, 0) can be converted into a unit vector (1, 0, 0), the vector corresponding to unit B (0, 4, 0) can be converted into a unit vector (0, 1, 0), and the vector corresponding to unit C (0, 0, 5) can be converted into a unit vector (0, 0, 1). Assign the total weight: wtotal = 5 + 3 + 2 = 10, normalize the weight: unit A's weight is 5 / 10 = 0.5, unit B's weight is 3 / 10 = 0.3, and unit C's weight is 2 / 10 = 0.2. Weighted average to calculate the weighted sum: integrated vector = 0.5 x (1, 0, 0) + 0.3 x (0, 1, 0) + 0.2 x (0,

[0082] 0,1) = (0.5, 0.3, 0.2). Normalize the integrated vector to a unit vector: modulus ≈ 0.616, global diffusion direction vector ≈ (0.812, 0.487, 0.325).

[0083] Step 124, Fourier transform the water pressure fluctuation parameter, extract the phase feature in the specific frequency range matching the resonance frequency of the sealed cavity structure from the Fourier transform result, perform sliding window smoothing and low pass filtering on the phase feature in the specific frequency range, and generate the denoised phase feature. The denoised phase feature is taken as the phase feature of the water pressure fluctuation parameter.

[0084] Wherein, the Fourier transform is to convert the water pressure fluctuation parameter from time domain to frequency domain, which is convenient for extracting the phase feature. The resonance frequency of the sealed cavity structure refers to the vibration frequency inherent to the sealed cavity (such as pressure vessel, pipeline, etc. closed structure) due to its physical properties (such as geometric shape, material stiffness, boundary constraint, etc.) under external excitation (such as water pressure fluctuation, mechanical vibration, etc.). When the excitation frequency is consistent with the cavity resonance frequency, the cavity will resonate, showing a significant increase in amplitude, which may lead to structural fatigue and even damage.

[0085] The embodiment of the present application extracts a frequency component consistent with the resonant frequency of the sealed cavity structure from the Fourier transform result, and the component is used to reflect the influence of leakage on water pressure. The sliding window smoothing processing refers to reducing the fluctuation of the phase feature by sliding window averaging method. It should be understood that the sliding window averaging method is a data processing technology that reduces the influence of random error or noise by calculating the average value of multiple data points in the phase feature within a specific frequency range, thereby improving the stability and reliability of the denoised phase feature. The core idea is to use the statistical characteristics of the phase feature within a specific frequency range to offset the abnormal fluctuation of individual phase features, thereby obtaining a result closer to the true value.

[0086] Step 125, according to the phase feature of the water pressure fluctuation parameter, calculating the water pressure phase fluctuation amplitude of each grid unit in the corresponding time window.

[0087] Among them, the time window water pressure phase fluctuation amplitude reflects the change amplitude of the water pressure fluctuation in the phase, and is used to quantify the influence of leakage on water pressure. The time window refers to dividing the continuous time series data into fixed length time periods for localized analysis and calculation of dynamic change statistical characteristics.

[0088] In step 125, the time window is used to intercept the data segment of the water pressure fluctuation phase feature in a certain time period, and then calculate the phase fluctuation amplitude in the time period to quantify the influence of leakage on water pressure.

[0089] The following is a specific example of a time window: assuming that the phase feature data of the water pressure fluctuation parameter has a duration of 10 seconds, the sampling frequency is 100 Hz, that is, 100 data points per second, and there are a total of 1000 phase values, corresponding to the time sequence t=0, 0.01, 0.02, …, 9.99 seconds, the following time window parameters are selected: window length: 1 second, 100 data points in 1 second, sliding step length: 0.5 seconds, window moves forward by 0.5 seconds each time, overlap rate: 50%, the first window is divided by the time window: t=0-1 second, containing the first 100 data points (index 0-99), the second window: t=0.5-1.5 seconds, containing data point index 50-149, the third window: t=1-2 seconds, data point index 100-199, and so on, until all 10 seconds of data are covered.

[0090] Step 126, based on the helium concentration gradient value and the water pressure phase fluctuation amplitude of each grid unit, a three-dimensional coupling coefficient matrix is constructed, and a tensor product operation is performed on the three-dimensional coupling coefficient matrix and the global diffusion direction vector to generate a target composite data frame.

[0091] Wherein, the three-dimensional coupling coefficient matrix reflects the coupling relationship between the helium concentration gradient value and the water pressure phase fluctuation amplitude. The target composite data frame generated by the tensor product operation is used for leakage monitoring and judgment.

[0092] The following is a specific example:

[0093] In the automatic sprinkler fire extinguishing system of a certain chemical plant, the rain alarm valve group detects abnormal helium concentration in the sealed cavity. The remote monitoring system first performs noise suppression and three-dimensional gridding processing on the helium concentration distribution data, identifies the initial helium concentration abnormal diffusion area; determine the target helium concentration abnormal diffusion area and its spatial coordinates through region growing algorithm; further, calculate the global diffusion direction vector. At the same time, the remote monitoring system performs Fourier transform and filtering processing on the water pressure fluctuation parameters, extracts the phase feature and calculates the water pressure phase fluctuation amplitude; finally, construct a three-dimensional coupling coefficient matrix, and perform tensor product operation with the global diffusion direction vector to generate a target composite data frame for subsequent leakage monitoring and control.

[0094] By performing steps 121-126, the scheme of the embodiment of the application realizes accurate analysis of the helium concentration abnormal diffusion area and the water pressure fluctuation feature through noise suppression, region growing, data fusion and tensor operation, generates a target composite data frame, provides reliable data support for leakage monitoring and control, improves the accuracy and real-time performance of the rain alarm valve group leakage monitoring, and is suitable for the demand of automatic sprinkler fire extinguishing system under complex working conditions.

[0095] In one possible embodiment, step 126, based on the helium concentration gradient value and the water pressure phase fluctuation amplitude of each grid cell, constructs a three-dimensional coupling coefficient matrix, and performs tensor product operation on the three-dimensional coupling coefficient matrix and the global diffusion direction vector to generate a target composite data frame, including:

[0096] Step a1, normalize the helium concentration gradient value of each grid cell respectively to generate a sequence of normalized helium concentration gradient values, and standardize the water pressure phase fluctuation amplitude of each grid cell in the corresponding time window to generate a sequence of standardized water pressure phase fluctuation amplitudes.

[0097] Wherein, the normalization processing scales the helium concentration gradient value to a unified range (such as [0, 1]), eliminates the dimensional difference, and facilitates subsequent calculation. The standardization processing converts the water pressure phase fluctuation amplitude into a distribution with a mean value of 0 and a standard deviation of 1, improving data comparability. The sequence of normalized helium concentration gradient values is a sequence formed by arranging the normalized helium concentration gradient values according to the grid cells. The sequence of standardized water pressure phase fluctuation amplitudes is a sequence formed by arranging the standardized water pressure phase fluctuation amplitudes according to the grid cells.

[0098] Step a2, based on the sequence of helium concentration gradient values and the sequence of water pressure phase fluctuation amplitudes, calculate the gradient phase coupling coefficient of each grid cell, which is obtained by multiplying the helium concentration gradient value and the water pressure phase fluctuation amplitude.

[0099] Wherein, the gradient phase coupling coefficient reflects the coupling relationship between the helium concentration gradient value and the water pressure phase fluctuation amplitude, and is used to quantify the influence of leakage on water pressure. Coupling is to couple the gradient and the phase.

[0100] Step a3, arrange the gradient phase coupling coefficients of all grid cells according to the spatial position to construct a three-dimensional coupling coefficient matrix, and perform tensor expansion on the global diffusion direction vector to generate a diffusion direction tensor.

[0101] Step a4, combine the three-dimensional coupling coefficient matrix and the diffusion direction tensor to perform element-by-element multiplication to generate an initial composite data frame.

[0102] Wherein, the three-dimensional coupling coefficient matrix is a three-dimensional matrix formed by arranging the gradient phase coupling coefficients according to the spatial position of the grid cells. The preliminary fusion data generated by the element-by-element multiplication operation is used for subsequent processing. The tensor is converted into a structure consistent with the dimension of the three-dimensional coupling coefficient matrix by tensor expansion of the global diffusion direction vector.

[0103] Step a5, perform spatial smoothing processing on the initial composite data frame to eliminate local data mutations and generate a target composite data frame.

[0104] Wherein, the spatial smoothing processing eliminates local mutations in the initial composite data frame through a filtering algorithm to improve data smoothness. The target composite data frame is the final fusion data after smoothing processing, which is used for leakage monitoring and judgment.

[0105] The following is a specific example:

[0106] In the automatic sprinkler fire extinguishing system of a large commercial complex, helium concentration anomaly is detected in the sealed cavity of the deluge alarm valve group. The system first normalizes and standardizes the helium concentration gradient value and the water pressure phase fluctuation amplitude to generate the corresponding sequence; further, the gradient phase coupling coefficient of each grid cell is calculated, and a three-dimensional coupling coefficient matrix is constructed; at the same time, the global diffusion direction vector is tensor expanded to generate a diffusion direction tensor; the initial composite data frame is generated by element-by-element multiplication operation, and spatial smoothing processing is performed to finally generate the target composite data frame, which is used for leakage monitoring and control.

[0107] By performing steps a1-a5, the present application realizes high-precision fusion of helium concentration gradient value and water pressure phase fluctuation amplitude through normalization, standardization, coupling coefficient calculation, tensor operation and spatial smoothing processing, generates a target composite data frame, provides a reliable data basis for leakage monitoring, improves the accuracy and real-time performance of leakage judgment, and is suitable for the needs of automatic water spray extinguishing systems under complex working conditions.

[0108] In one possible embodiment, step a3 combines the three-dimensional coupling coefficient matrix with the diffusion direction tensor to perform element-by-element multiplication operation to generate an initial composite data frame, including:

[0109] Step b1 performs dimension checking on the three-dimensional coupling coefficient matrix to ensure that the spatial dimensions of the three-dimensional coupling coefficient matrix and the diffusion direction tensor are consistent.

[0110] The dimension checking checks whether the spatial dimensions of the three-dimensional coupling coefficient matrix are consistent with the diffusion direction tensor to ensure data compatibility. The spatial dimensions being consistent means that the size (such as length, width, and height) of the matrix and the tensor in the three-dimensional space completely matches.

[0111] Step b2 performs weight correction on each element in the three-dimensional coupling coefficient matrix to adjust the weight value of the gradient phase coupling coefficient, generates a corrected three-dimensional coupling coefficient matrix, and performs normalization processing on the diffusion direction tensor to ensure that the module length of the diffusion direction vector of each representative grid cell is a unit length, and generates a normalized diffusion direction tensor.

[0112] The weight correction adjusts the weight value of the gradient phase coupling coefficient according to a preset rule to optimize the data fusion effect. The corrected three-dimensional coupling coefficient matrix is the gradient phase coupling coefficient matrix after weight correction. The normalized diffusion direction tensor is the diffusion direction tensor with a module length of a unit length.

[0113] Step b3 performs element-by-element multiplication operation on the corrected three-dimensional coupling coefficient matrix and the normalized diffusion direction tensor to generate an initial composite data frame.

[0114] The element-by-element multiplication operation means multiplying the elements at corresponding positions in the matrix and the tensor to generate a new data structure. The initial composite data frame is the preliminary fusion data generated by the element-by-element multiplication operation, which is used for subsequent processing.

[0115] The following is a specific example:

[0116] In the automatic water spray fire extinguishing system of a certain chemical plant, helium concentration anomaly is detected in the sealing cavity of the deluge alarm valve group. The system first checks the dimension of the three-dimensional coupling coefficient matrix to ensure that it is consistent with the spatial dimension of the diffusion direction tensor; then, the weight of each element in the three-dimensional coupling coefficient matrix is corrected to generate a corrected three-dimensional coupling coefficient matrix, and the diffusion direction tensor is normalized; finally, the corrected three-dimensional coupling coefficient matrix and the normalized diffusion direction tensor are multiplied element by element to generate an initial composite data frame, which is used for leakage monitoring and control.

[0117] By performing steps b1-b3, the present embodiment of the application realizes high-precision fusion of the three-dimensional coupling coefficient matrix and the diffusion direction tensor through dimension checking, weight correction, normalization processing and element-by-element multiplication operation, generates an initial composite data frame, provides a more reliable data basis for leakage monitoring, improves the accuracy and real-time performance of leakage judgment, and is suitable for the demand of automatic water spray fire extinguishing system under complex working conditions.

[0118] In a possible embodiment, step a5, the initial composite data frame is subjected to spatial smoothing processing to eliminate local data mutations and generate a target composite data frame, comprising:

[0119] Step a51, local neighborhood analysis is performed on the helium concentration gradient value of each grid cell in the initial composite data frame. In the local neighborhood analysis process, the difference in helium concentration gradient value between each grid cell and its adjacent grid cell is calculated to generate a local data difference matrix.

[0120] Wherein, the local neighborhood analysis refers to the comparison and analysis of the helium concentration gradient value of each grid cell and its adjacent cell, and the identification of local data difference. The local data difference matrix records the difference in helium concentration gradient value between each grid cell and its adjacent cell, which is used to identify the local mutation area.

[0121] Step a52, based on the local data difference matrix, identify the grid cells whose helium concentration gradient value difference exceeds the preset difference threshold, and mark them as local mutation grid cells. The helium concentration gradient value of the local mutation grid cells is corrected by interpolation to generate a corrected helium concentration gradient value.

[0122] Wherein, step a52 marks the grid cells whose helium concentration gradient value difference exceeds the preset difference threshold as local mutation grid cells. The grid cells whose helium concentration gradient value difference exceeds the preset threshold are likely to be data outliers. Interpolation correction refers to correcting the value of the local mutation grid cell based on the helium concentration gradient value of the adjacent grid cells to eliminate the anomaly.

[0123] For example, the local data difference matrix is [0.3, 0.8, 1.5, 0.7, 0.4], and the helium concentration gradient values of each unit are as follows: the helium concentration gradient value of unit 1 is 1.2, the helium concentration gradient value of unit 2 is 1.5, the helium concentration gradient value of unit 3 is 3.0, the helium concentration gradient value of unit 4 is 1.8, and the helium concentration gradient value of unit 5 is 1.3. There are 5 grid units. The interpolation correction is performed on the local mutation unit (such as unit 3). The directly adjacent units of the mutation unit are unit 2 (gradient value 1.5) on the left side of unit 3 and unit 4 (gradient value 1.8) on the right side of unit 3. The correction value is calculated using the arithmetic mean method, and the helium concentration gradient value of unit 3 is corrected to generate the corrected helium concentration gradient value.

[0124] Step a53, updating the corrected helium concentration gradient value to the initial composite data frame to generate an intermediate composite data frame, performing smoothing processing on the intermediate composite data frame, adjusting the helium concentration gradient value of each grid unit based on a preset spatial continuity constraint to generate a smoothed composite data frame.

[0125] The intermediate composite data frame is the data frame after updating the corrected helium concentration gradient value, which is used for further processing. The smoothing processing eliminates local fluctuations in the data by using a filtering algorithm to improve the smoothness of the data. The spatial continuity constraint ensures that the change in the helium concentration gradient value of adjacent grid units conforms to the physical law.

[0126] Step a54, performing boundary consistency verification on the smoothed composite data frame, correcting abnormal data points at the boundary of each local mutation grid unit based on the gradient change of the data value of adjacent grid units to generate a target composite data frame.

[0127] The boundary consistency verification checks whether the data value at the boundary of the local mutation grid unit conforms to the gradient change rule of the adjacent unit.

[0128] The following is a specific example:

[0129] In the automatic sprinkler fire extinguishing system of a high-rise building, helium concentration anomalies are detected in the sealed cavity of the deluge alarm valve group. The system first performs local neighborhood analysis on the initial composite data frame to generate a local data difference matrix. Then, the helium concentration gradient value of the local mutation grid unit is identified and corrected to generate an intermediate composite data frame. Further, the intermediate composite data frame is smoothed to generate a smoothed composite data frame. Finally, the boundary consistency verification is performed on the smoothed composite data frame to correct the abnormal data points and generate a target composite data frame for leak monitoring and control.

[0130] By performing steps a51-a54, the scheme of the embodiment of the application realizes high-precision correction and optimization of the initial composite data frame through local neighborhood analysis, interpolation correction, smoothing processing and boundary consistency verification, generates the target composite data frame, and improves the accuracy and reliability of the leakage monitoring data, which is suitable for the demand of automatic water-spraying fire extinguishing system under complex working conditions.

[0131] In a possible embodiment, S13, in combination with the spatial coordinates and the global diffusion direction vector, determines the global leakage rate gradient of the target composite data frame. When the global leakage rate gradient exceeds the preset reference curve for multiple times in succession, the multiple harmonic distortion amplification quantities in the current phase feature are extracted, and the leakage level identifier is determined according to the multiple harmonic distortion amplification quantities and the global leakage rate gradient, including:

[0132] Step c1, determine the global leakage rate gradient of each grid cell based on the spatial coordinates, accumulate the leakage rate gradient of each grid cell in time sequence to generate a leakage rate gradient time sequence, and determine the global leakage rate gradient of the target composite data frame based on the leakage rate gradient time sequence and in combination with the global diffusion direction vector.

[0133] The global leakage rate gradient can reflect the change trend of the leakage rate in space and time, and is used to quantify the leakage severity. The leakage rate gradient time sequence refers to a sequence formed by arranging the leakage rate gradient of each grid cell in time sequence. The global leakage rate gradient of the target composite data frame is the final leakage rate gradient generated by comprehensively considering the time sequence and the global diffusion direction vector. The global leakage rate gradient refers to the trend of the helium concentration in each cell inside the sealed cavity changing with time after the cavity is divided into multiple uniformly distributed grid cells. This trend reflects the speed of helium gas (as a tracer gas) leaking in a particular region, and by analyzing these gradient values, the specific location and diffusion direction of the leakage can be located. Specifically, the global leakage rate gradient is constructed based on the helium concentration gradient value in the grid cell and the water pressure phase fluctuation amplitude, and reflects the dynamic leakage in the local area.

[0134] Step c2, determine whether the global leakage rate gradient exceeds the preset reference curve for multiple times in succession.

[0135] The preset reference curve is a leakage rate gradient threshold value set according to historical data or experimental data, and is used to determine whether the leakage is abnormal.

[0136] Step c3, if yes, determine the abnormal time window, which is the time window corresponding to the situation that the global leakage rate gradient exceeds the preset reference curve for multiple times in succession.

[0137] The abnormal time window refers to a time period in which the global leakage rate gradient continuously exceeds the preset reference curve multiple times, and is used for further analysis of the leakage situation. Multiple times can refer to 3 times, 4 times, 10 times, etc. The specific value of multiple times is not limited in the embodiments of the present application.

[0138] Step c4, in the abnormal time window, the multiple harmonic distortion amplification quantity in the current phase feature is extracted, and the multiple harmonic distortion amplification quantity is time-integrated to generate a harmonic distortion cumulative quantity.

[0139] The harmonic distortion cumulative quantity refers to the total quantity generated by time-integrating the multiple harmonic distortion amplification quantity, and is used to quantify the influence of leakage on the current phase.

[0140] Step c5, based on the ratio relationship between the harmonic distortion cumulative quantity and the global leakage rate gradient, the leakage level identifier is determined.

[0141] The leakage level identifier refers to the leakage severity level determined according to the ratio relationship between the harmonic distortion cumulative quantity and the global leakage rate gradient.

[0142] The following is a specific example:

[0143] In the automatic sprinkler fire extinguishing system of a certain chemical plant, helium concentration anomaly is detected in the sealing cavity of the deluge alarm valve group. The remote monitoring system first calculates the leakage rate gradient of each grid unit based on the spatial coordinates, and generates a leakage rate gradient time series; further, in combination with the global diffusion direction vector, the global leakage rate gradient of the target composite data frame is determined. Subsequently, when the remote monitoring system determines that the global leakage rate gradient exceeds the preset reference curve continuously three times, the abnormal time window is determined; in the abnormal time window, the multiple harmonic distortion amplification quantity in the current phase feature is extracted, and the multiple harmonic distortion amplification quantity is time-integrated to generate a harmonic distortion cumulative quantity; finally, based on the ratio relationship between the harmonic distortion cumulative quantity and the global leakage rate gradient, the leakage level is determined to be "severe", and the corresponding leakage level identifier is generated.

[0144] By performing steps c1-c5, the embodiments of the present application realize accurate monitoring and evaluation of the leakage situation through leakage rate gradient time series analysis, abnormal time window determination, harmonic distortion cumulative quantity calculation and leakage level determination, improve the accuracy and real-time performance of the leakage judgment, and are suitable for the demand of automatic sprinkler fire extinguishing system under complex working conditions.

[0145] In a possible embodiment, step c1, based on the leakage rate gradient time series, in combination with the global diffusion direction vector, determines the global leakage rate gradient of the target composite data frame, comprising:

[0146] Step d1, calculate the helium concentration gradient value change rate between grid cells according to the helium concentration gradient value of each grid cell.

[0147] Wherein, the helium concentration gradient value change rate reflects the degree of change of the helium concentration gradient value between adjacent grid cells, which is used to quantify the dynamic characteristics of leakage diffusion.

[0148] Step d2, calculate the weight of each grid cell according to the helium concentration gradient value change rate, combined with the global diffusion direction vector, the distance from the grid cell to the leakage source, and the estimated diffusion speed of helium in the target helium concentration anomaly diffusion area.

[0149] Wherein, the grid cell weight reflects the importance of each grid cell in the calculation of the leakage rate gradient, considering factors such as gradient value change rate, diffusion direction, distance, and diffusion speed.

[0150] Step d3, optimize the weight of each grid cell by combining historical data to obtain the optimized weight of each grid cell, and construct a weighted matrix.

[0151] Wherein, the historical data is historical leakage data, and step d3 uses historical leakage data to correct the weight to improve the accuracy of weight calculation. The weighted matrix refers to the matrix formed by arranging the optimized weight according to the position of the grid cell, which is used for subsequent leakage rate gradient calculation. Grid cells closer to the leakage source may have higher weights, because grid cells closer to the leakage source more directly reflect the immediate state of the leakage; while areas with faster diffusion speed may also obtain higher weights, because areas with faster diffusion speed indicate that the leaked material can spread faster.

[0152] Step d4, combine the leakage rate gradient time series with the weighted matrix to obtain the global leakage rate gradient of the target composite data frame.

[0153] Wherein, the global leakage rate gradient refers to the final leakage rate gradient generated by combining the leakage rate gradient time series and the weighted matrix, which is used to quantify the global trend of the leakage.

[0154] The following is a specific example:

[0155] In the automatic sprinkler fire extinguishing system of a certain chemical plant, helium concentration anomaly is detected in the sealed cavity of the deluge alarm valve group. The remote monitoring system first calculates the helium concentration gradient value change rate of each grid cell; then, combined with the global diffusion direction vector, the distance from the grid cell to the leakage source, and the estimated diffusion speed of helium, the weight of each grid cell is calculated; further, the weight is optimized using historical data to construct a weighted matrix; finally, the leakage rate gradient time series is combined with the weighted matrix to generate the global leakage rate gradient of the target composite data frame, which is used for leakage monitoring and control.

[0156] By performing steps d1-d4, the scheme of the embodiment of the application realizes high-precision calculation of the leakage rate gradient by helium concentration gradient value change rate calculation, weight optimization and weighted matrix construction, generates the global leakage rate gradient of the target composite data frame, and improves the accuracy and reliability of the leakage monitoring, which is suitable for the demand of automatic sprinkler fire extinguishing system under complex working conditions.

[0157] Figure 2 A structural diagram of a valve group operation state remote monitoring system provided by the embodiment of the application is shown in FIG. Figure 2 The system comprises:

[0158] The acquisition module 21 acquires the helium concentration distribution data inside the sealed cavity of the deluge alarm valve group, the water pressure fluctuation parameters inside the deluge alarm valve group, and the current phase characteristics of the fire pump station, which is taken as a water supply power source and establishes a protocol cascade with the deluge alarm valve group.

[0159] The construction module 22 determines the spatial coordinates and global diffusion direction vector of the target helium concentration abnormal diffusion area inside the sealed cavity based on the helium concentration distribution data, and constructs the target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data.

[0160] The extraction module 23 determines the global leakage rate gradient of the target composite data frame in combination with the spatial coordinates and global diffusion direction vector, extracts the multiple harmonic distortion amplification quantities in the current phase characteristics when the global leakage rate gradient continuously exceeds the preset reference curve for multiple times, and determines the leakage level identifier according to the multiple harmonic distortion amplification quantities and the global leakage rate gradient.

[0161] The generation module 24 generates a linkage instruction containing the leakage level identifier, the spatial coordinates and the global diffusion direction vector, and triggers the operation state closed-loop control process of the deluge alarm valve group through the linkage instruction.

[0162] Figure 2 The rain alarm valve group operation state remote monitoring system can perform Figure 1 The rain alarm valve group operation state remote monitoring method is not described again in detail. The specific operation of each module and unit of the rain alarm valve group operation state remote monitoring system in the above embodiment has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0163] In one possible design, Figure 2 The rain alarm valve group operation state remote monitoring system in the embodiment shown in FIG. Figure 3 The computing device can comprise a storage component 31 and a processing component 32.

[0164] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0165] The processing component 32 is configured to: acquire helium concentration distribution data inside a sealed cavity of a deluge alarm valve group, water pressure fluctuation parameters inside the deluge alarm valve group, and current phase characteristics of a fire pump station, the fire pump station being a water supply power source and establishing a protocol cascade with the deluge alarm valve group; determine spatial coordinates of a target helium concentration abnormal diffusion area inside the sealed cavity and a global diffusion direction vector based on the helium concentration distribution data, and construct a target composite data frame based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data; determine a global leakage rate gradient of the target composite data frame in combination with the spatial coordinates and the global diffusion direction vector, extract a multiple harmonic distortion amplification amount in the current phase characteristics when the global leakage rate gradient exceeds a preset reference curve for multiple times in succession, and determine a leakage level identifier according to the multiple harmonic distortion amplification amount and the global leakage rate gradient; and generate a linkage instruction containing the leakage level identifier, the spatial coordinates, and the global diffusion direction vector, and trigger a running state closed-loop control process of the deluge alarm valve group through the linkage instruction.

[0166] The processing component 32 can include one or more processors to execute the computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors, or other electronic elements for executing the above method.

[0167] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as a random access memory (RAM), a static random access memory (SRAM), an electrically erasable programmable read only memory (EEPROM), an erasable programmable read only memory (EPROM), a programmable read only memory (PROM), a read only memory (ROM), a magnetic storage, a flash memory, a magnetic disk, or a optical disk.

[0168] Of course, the computing device can also include other components, such as an input / output interface, a display component, a communication component, etc.

[0169] The input / output interface provides an interface between the processing component and peripheral interface modules, which can be output devices, input devices, etc.

[0170] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0171] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform, and the processing component, the storage component, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0172] The embodiments of the present application also provide a computer storage medium storing a computer program, and the computer program can implement the above-mentioned Figure 1 A rain shower alarm valve group operation state remote monitoring method is provided in the embodiments.

[0173] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, device and unit can refer to the corresponding process in the foregoing method embodiments, and will not be described here.

[0174] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0175] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0176] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for remotely monitoring the operating status of a rain alarm valve assembly, characterized in that, include: Data on helium concentration distribution inside the sealed cavity of the deluge alarm valve assembly, water pressure fluctuation parameters inside the deluge alarm valve assembly, and current phase characteristics of the fire pump station are obtained. The fire pump station serves as a water supply power source and establishes a protocol-level linkage with the deluge alarm valve assembly. Based on the helium concentration distribution data, the spatial coordinates and global diffusion direction vector of the target helium concentration anomaly diffusion region inside the sealed cavity are determined. Based on the phase characteristics of the water pressure fluctuation parameters and the helium concentration distribution data, a target composite data frame is constructed. Among them, the helium concentration distribution data is subjected to noise suppression processing to generate denoised helium concentration distribution data. The denoised helium concentration distribution data is then three-dimensionally meshed according to the geometry of the sealed cavity to generate uniformly distributed mesh cells. The helium concentration gradient value of each mesh cell is calculated. The water pressure fluctuation parameters are subjected to Fourier transform, and phase features within a specific frequency range that match the resonant frequency of the sealed cavity structure are extracted from the Fourier transform results. The phase features within the specific frequency range are then subjected to sliding window smoothing and low-pass filtering to generate denoised phase features. The denoised phase features are then used as the phase features of the water pressure fluctuation parameters. Based on the phase characteristics of the water pressure fluctuation parameters, the amplitude of the water pressure phase fluctuation of each grid cell within the corresponding time window is calculated. Based on the helium concentration gradient value of each grid cell and the water pressure phase fluctuation amplitude, a three-dimensional coupling coefficient matrix is ​​constructed, and the three-dimensional coupling coefficient matrix is ​​multiplied by the global diffusion direction vector to generate a target composite data frame. Combining the spatial coordinates and the global diffusion direction vector, the global leakage rate gradient of the target composite data frame is determined. When the global leakage rate gradient exceeds the preset reference curve multiple times in a row, the multiple harmonic distortion amplification amount in the current phase characteristics is extracted. Based on the multiple harmonic distortion amplification amount and the global leakage rate gradient, the leakage level identifier is determined. Generate a linkage command that includes the leakage level identifier, the spatial coordinates, and the global diffusion direction vector, and trigger the closed-loop control process of the deluge alarm valve assembly's operating status through the linkage command.

2. The method according to claim 1, characterized in that, The method further includes: Identify regions of anomalous initial helium concentration diffusion; A region growth process is performed on the initial helium concentration abnormal diffusion region. During the region growth process, adjacent mesh cells are merged based on a preset gradient direction consistency criterion until the gradient value at the growth boundary is lower than a preset gradient threshold, thereby obtaining the target helium concentration abnormal diffusion region and determining the spatial coordinates of the target helium concentration abnormal diffusion region. Within the target helium concentration anomalous diffusion region, several grid cells with larger gradient values ​​are selected as representative grid cells. The diffusion direction vector of each representative grid cell is calculated. All the diffusion direction vectors are then normalized and weighted to generate the global diffusion direction vector of the target helium concentration anomalous diffusion region.

3. The method according to claim 2, characterized in that, The process involves constructing a three-dimensional coupling coefficient matrix based on the helium concentration gradient values ​​of each grid cell and the amplitude of the water pressure phase fluctuation, and then performing a tensor product operation between the three-dimensional coupling coefficient matrix and the global diffusion direction vector to generate a target composite data frame, including: The helium concentration gradient values ​​of each grid cell are normalized to generate a normalized helium concentration gradient value sequence. The water pressure phase fluctuation amplitude of each grid cell within the corresponding time window is standardized to generate a standardized water pressure phase fluctuation amplitude sequence. Based on the helium concentration gradient value sequence and the water pressure phase fluctuation amplitude sequence, the gradient phase coupling coefficient of each grid cell is calculated. The gradient phase coupling coefficient is obtained by multiplying the helium concentration gradient value and the water pressure phase fluctuation amplitude. Arrange the gradient phase coupling coefficients of all mesh elements according to their spatial positions to construct a three-dimensional coupling coefficient matrix, and perform tensor expansion on the global diffusion direction vector to generate a diffusion direction tensor; By combining the three-dimensional coupling coefficient matrix with the diffusion direction tensor, an element-wise multiplication operation is performed to generate an initial composite data frame; The initial composite data frame is spatially smoothed to eliminate local data abrupt changes, thereby generating the target composite data frame.

4. The method according to claim 3, characterized in that, The step of combining the three-dimensional coupling coefficient matrix and the diffusion direction tensor, and performing element-wise multiplication to generate an initial composite data frame includes: The dimension of the three-dimensional coupling coefficient matrix is ​​verified to ensure that the spatial dimension of the three-dimensional coupling coefficient matrix is ​​consistent with that of the diffusion direction tensor. Each element in the three-dimensional coupling coefficient matrix is ​​weighted to adjust the weight value of the gradient phase coupling coefficient, generating a corrected three-dimensional coupling coefficient matrix. The diffusion direction tensor is normalized to ensure that the magnitude of the diffusion direction vector of each representative grid cell is a unit length, generating a normalized diffusion direction tensor. The modified three-dimensional coupling coefficient matrix is ​​multiplied element-wise with the normalized diffusion direction tensor to generate an initial composite data frame.

5. The method according to claim 3, characterized in that, The step of spatially smoothing the initial composite data frame to eliminate local data abrupt changes and generate the target composite data frame includes: Local neighborhood analysis is performed on the helium concentration gradient value of each grid cell in the initial composite data frame. During the local neighborhood analysis, the difference in helium concentration gradient value between each grid cell and its neighboring grid cells is calculated to generate a local data difference matrix. Based on the local data difference matrix, grid cells whose helium concentration gradient value difference exceeds a preset difference threshold are identified and marked as local mutation grid cells. The helium concentration gradient values ​​of the local mutation grid cells are then interpolated and corrected to generate corrected helium concentration gradient values. The corrected helium concentration gradient value is updated to the initial composite data frame to generate an intermediate composite data frame. The intermediate composite data frame is then smoothed by adjusting the helium concentration gradient value of each grid cell based on a preset spatial continuity constraint to generate a smoothed composite data frame. The smoothed composite data frame is subjected to boundary consistency verification. Based on the gradient change of data values ​​of adjacent grid cells, the abnormal data points at the boundary of each locally abrupt grid cell are corrected to generate the target composite data frame.

6. The method according to claim 1, characterized in that, The global leakage rate gradient of the target composite data frame is determined by combining the spatial coordinates and the global diffusion direction vector. When the global leakage rate gradient exceeds a preset reference curve multiple times consecutively, the multiple harmonic distortion amplification in the current phase characteristics is extracted. Based on the multiple harmonic distortion amplification and the global leakage rate gradient, a leakage level identifier is determined, including: The global leakage rate gradient of each grid cell is determined based on the spatial coordinates. The leakage rate gradient of each grid cell is accumulated over time to generate a leakage rate gradient time series. Based on the leakage rate gradient time series and combined with the global diffusion direction vector, the global leakage rate gradient of the target composite data frame is determined. Determine whether the global leakage rate gradient exceeds the preset reference curve multiple times consecutively; If so, then an abnormal time window is determined, which is the time window corresponding to when the global leakage rate gradient exceeds the preset benchmark curve multiple times consecutively. Within the abnormal time window, the multiple harmonic distortion amplification amount in the current phase characteristics is extracted, and the multiple harmonic distortion amplification amount is integrated over time to generate the harmonic distortion accumulation amount. The leakage level identifier is determined based on the ratio of the accumulated harmonic distortion to the global leakage rate gradient.

7. The method according to claim 6, characterized in that, The step of determining the global leakage rate gradient of the target composite data frame based on the leakage rate gradient time series and combined with the global diffusion direction vector includes: Calculate the rate of change of helium concentration gradient between grid cells based on the helium concentration gradient values ​​of each grid cell; Based on the rate of change of helium concentration gradient, combined with the global diffusion direction vector, the distance from the grid cell to the leakage source, and the estimated diffusion rate of helium in the target helium concentration anomalous diffusion area, the weight of each grid cell is calculated. By combining historical data, the weights of each grid cell are optimized to obtain the optimized weights of each grid cell, and a weighted matrix is ​​constructed. The global leakage rate gradient of the target composite data frame is obtained by combining the leakage rate gradient time series with the weighting matrix.

8. A remote monitoring system for the operating status of a rain alarm valve assembly, characterized in that, include: The acquisition module acquires helium concentration distribution data inside the sealed cavity of the deluge alarm valve group, water pressure fluctuation parameters inside the deluge alarm valve group, and current phase characteristics of the fire pump station. The fire pump station, as the water supply power source, establishes a protocol-level linkage with the deluge alarm valve group. The module constructs a target composite data frame based on helium concentration distribution data to determine the spatial coordinates and global diffusion direction vector of the target helium concentration anomaly diffusion region inside the sealed cavity. Based on the phase characteristics of water pressure fluctuation parameters and helium concentration distribution data, the module performs noise suppression processing on the helium concentration distribution data to generate denoised helium concentration distribution data. The denoised helium concentration distribution data is then meshed in three-dimensional space according to the geometry of the sealed cavity to generate uniformly distributed mesh cells. The helium concentration gradient value of each mesh cell is then calculated. The water pressure fluctuation parameters are subjected to Fourier transform, and phase features within a specific frequency range that match the resonant frequency of the sealed cavity structure are extracted from the Fourier transform results. The phase features within the specific frequency range are then subjected to sliding window smoothing and low-pass filtering to generate denoised phase features. The denoised phase features are then used as the phase features of the water pressure fluctuation parameters. Based on the phase characteristics of the water pressure fluctuation parameters, the amplitude of the water pressure phase fluctuation of each grid cell within the corresponding time window is calculated. Based on the helium concentration gradient value of each grid cell and the water pressure phase fluctuation amplitude, a three-dimensional coupling coefficient matrix is ​​constructed, and the three-dimensional coupling coefficient matrix is ​​multiplied by the global diffusion direction vector to generate a target composite data frame. The extraction module, combining spatial coordinates and global diffusion direction vector, determines the global leakage rate gradient of the target composite data frame. When the global leakage rate gradient exceeds the preset reference curve multiple times in a row, the multiple harmonic distortion amplification in the current phase characteristics is extracted. Based on the multiple harmonic distortion amplification and the global leakage rate gradient, the leakage level identifier is determined. The generation module generates linkage commands that include leakage level identifiers, spatial coordinates, and global diffusion direction vectors. These linkage commands trigger the closed-loop control process of the deluge alarm valve assembly's operating status.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a remote monitoring method for the operating status of a rain alarm valve group as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The device contains a computer program, which, when executed by a computer, implements a remote monitoring method for the operating status of a rain alarm valve assembly as described in any one of claims 1 to 7.

Images