A flow monitoring device based on a wireless intelligent monitoring ring

Abstract

The application discloses a kind of flow monitoring devices based on wireless intelligent monitoring ring, it is related to flow monitoring device technical field, to realize the accurate real-time monitoring of the water discharge of fire gun;The application includes wireless intelligent monitoring ring and back-end server, wherein wireless intelligent monitoring ring is equipped with annular shell, annular shell is provided with electronic compartment and measuring hole, sensor is placed in measuring hole, microprocessor is placed in electronic compartment, can real-time measurement water flow and environmental parameter data;Data is compensated and calibrated by data processing unit, and is transmitted to back-end server, and server calculates the water discharge of water gun and the residual water volume of vehicle-mounted water tank according to data, and adjusts the water discharge of water gun;The application improves the accuracy of water volume evaluation, improves the efficiency of fire-fighting action, and the device design is suitable for complex fire scene, enhances the adaptability and reliability of equipment;In addition, the application helps to avoid water resource waste, optimizes fire-fighting decision, improves fire-fighting success rate.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of flow monitoring devices, in particular, relates to a flow monitoring device based on a wireless intelligent monitoring ring. BACKGROUND

[0002] When carrying out fire fighting, the fire fighting strategy is determined according to the fire situation. For larger or more complex fire situations, multiple water guns need to be connected to the vehicle-mounted water tank of the fire truck, and multiple branches are branched out from the main water supply pipeline to supply water to each water gun. During this period, the problem of continuous water supply needs to be considered. When the remaining capacity of the vehicle-mounted water tank is insufficient to extinguish the fire, remedial measures are taken as soon as possible to ensure sufficient water for fire fighting.

[0003] The water output of the branches branched out from the main water supply pipeline is not the same, and the remaining water volume of the vehicle-mounted water tank needs to be evaluated by monitoring the water output of each branch. At present, the firemen estimate the water output of the water gun they control according to their feelings and experience, and then communicate with each other to estimate the remaining water volume of the vehicle-mounted water tank. This method is inefficient and cannot accurately estimate the remaining water volume of the vehicle-mounted water tank, affecting subsequent fire fighting decisions. SUMMARY

[0004] (I) Technical problems solved

[0005] To solve the problems in the related art, the application provides a flow monitoring device based on a wireless intelligent monitoring ring. The application solves the technical problems of inaccurate and inefficient evaluation of the remaining water volume of the vehicle-mounted water tank by combining real-time data acquisition by sensors, data processing by microprocessors, and calculation and analysis by back-end servers, and realizes accurate monitoring of the water output of each water gun and accurate estimation of the remaining water volume of the vehicle-mounted water tank, thereby improving the scientificity of fire fighting decisions and the effectiveness of resource management.

[0006] (II) Technical solutions

[0007] To solve the above technical problems, the application is implemented by the following technical solutions:

[0008] A flow monitoring device based on a wireless intelligent monitoring ring, characterized in that it comprises a wireless intelligent monitoring ring and a back-end server.

[0009] The wireless intelligent monitoring ring comprises a circular shell 2 arranged between the water gun and the water hose, an electronic compartment 1 and a measuring hole 3 arranged on the circular shell 2.

[0010] The two ends of the circular shell 2 are provided with quick connection devices, and the connection parts are provided with sealing rings.

[0011] The surface of the shell is designed with an impact-resistant protective layer, and the inner wall is provided with a shockproof structure.

[0012] The casing is designed with ventilation holes to ensure that the device can still work normally in high-temperature environments;

[0013] The sensor, placed in the measuring hole 3, is used to measure the water flow parameters and environmental parameters of the water hose and water gun in real time.

[0014] The microprocessor, placed in the electronic compartment 1, includes a signal acquisition unit, a data processing unit, and a water volume calculation unit;

[0015] The signal acquisition unit is connected to the sensor via leads and is used to acquire water flow data and environmental data near the water gun.

[0016] The data processing unit is used to compensate and calibrate the water flow data based on environmental data.

[0017] The data transmission unit is used to send the compensated and calibrated water flow data to the backend server.

[0018] The backend server is used to calculate the water output of each water gun and the remaining water volume in the vehicle's water tank based on the compensated and calibrated water flow data; and to adjust the water output of the water gun based on the water output of each water gun and the remaining water volume in the vehicle's water tank.

[0019] Preferably, the sensor is placed in the measuring hole 3, and the specific steps for measuring the water flow parameters and environmental parameters flowing through the water hose and water gun in real time are as follows:

[0020] S11, Set the water flow parameter set Set environmental parameter set ,in a i Represents the first in the set of flow parameters i A water flow parameter, n This indicates the total number of flow parameters. b i Represents the first in the set of environmental parameters i One environmental parameter, q Indicates the total number of environmental parameters;

[0021] S12. Arrange sensors in the measuring holes 3 of the wireless intelligent monitoring ring according to the parameters in the water flow parameter set and the environmental parameter set.

[0022] The above steps, by setting water flow parameter sets and environmental parameter sets, and accordingly arranging sensors in the measuring holes 3 of the wireless intelligent monitoring ring, enable real-time and accurate measurement of water flow parameters and environmental parameter data flowing through the water hose and water gun, effectively improving the accuracy and comprehensiveness of data acquisition.

[0023] Preferably, the signal acquisition unit is connected with the sensor through a lead wire, and the specific steps for acquiring the water flow data and the environmental data near the water gun are as follows:

[0024] S21, collect the number of real-time water guns to obtain the number of real-time water guns as m , and m wireless intelligent monitoring rings are connected with the water gun and the water hose through the quick connection device at both ends of the circular ring-shaped shell 2.

[0025] The real-time water flow parameter data and the real-time environmental parameter data measured by the sensor through the water hose and the water gun are collected through the lead wire to obtain a real-time monitoring water flow parameter data matrix A and a real-time environmental data parameter matrix B , as follows,

[0026] ;

[0027] ;

[0028] wherein, A in represents the real-time water flow parameter data matrix A , the real-time water flow parameter data matrix i , the real-time water flow parameter data matrix n , the real-time water flow parameter data matrix B iq represents the real-time environmental parameter data matrix i , the real-time environmental parameter data matrix q , the real-time environmental parameter data matrix

[0029] The above steps can efficiently collect the real-time water flow parameters and environmental parameters of each water gun through the lead wire connected with the sensor to form a detailed data matrix, ensure the real-time and accuracy of data acquisition, realize the parallel monitoring of multiple water guns, and provide comprehensive and reliable data support for subsequent fluid mechanics model calculation and water volume evaluation.

[0030] Preferably, the data processing unit is used for compensating and calibrating the water flow data according to the environmental data, and the specific steps are as follows:

[0031] S31, collect the actual water flow parameter data under different historical environments and the water flow parameter data measured by the wireless intelligent monitoring ring to obtain a historical environmental and monitoring parameter data matrix and a historical actual water flow parameter data matrix, and take the historical actual water flow parameter data matrix as a label matrix of the historical environmental and monitoring parameter data matrix.

[0032] S32, construct an initial CNN neural network model, and set the initial weight of the initial CNN neural network model as c 1, the initial bias is d1. Initial learning rate e 1; Set the training ratio and testing ratio of the initial CNN neural network model as follows: f 1. f 2;

[0033] S33, according to the above f 1. f 2. Divide the historical environment and monitoring parameter data matrix into a historical environment and monitoring parameter data training matrix and a historical environment and monitoring parameter data test matrix; correspondingly, divide the historical actual water flow parameter data matrix into a historical actual water flow parameter data training matrix and a historical actual water flow parameter data test matrix.

[0034] S34. Input the training matrix of historical environment and monitoring parameter data and the training matrix of historical actual water flow parameter data into the initial CNN neural network model for training. After training, the trained CNN neural network model is obtained.

[0035] S35. Input the historical environment and monitoring parameter data test matrix and the historical actual water flow parameter data test matrix into the trained CNN neural network model for testing. After the test optimization is completed, the final CNN neural network model is obtained.

[0036] S36, Real-time monitoring of water flow parameter data matrix A and real-time environmental parameter data matrix B The data is fed into the final CNN neural network model to obtain the compensated and corrected real-time water flow parameter data matrix. C ,as follows,

[0037] ;

[0038] in, Cin The first element in the real-time flow parameter data matrix for compensation and correction is... i The water gun after compensation and calibration n Real-time water flow parameter data;

[0039] The above steps, by collecting historical data and constructing a label matrix, provide rich learning samples for neural network training, ensuring the effectiveness of the model. After training and testing optimization, the CNN neural network model can accurately compensate and correct the real-time monitored water flow parameter data, eliminating the influence of environmental factors on the measurement results. The final generated compensated and corrected real-time water flow parameter data matrix provides more accurate water volume information, which helps to achieve more refined water gun output adjustment and more efficient fire extinguishing strategy formulation.

[0040] Preferably, the specific steps of S34 are as follows:

[0041] S341, set a training error threshold and a maximum number of iterations;

[0042] S342, input the historical environment and monitoring parameter data training matrix and the historical actual water flow parameter data training matrix into an initial CNN neural network model, and repeatedly train the initial CNN neural network model;

[0043] S343, adjust the initial weight, initial bias and initial learning rate of the initial CNN neural network model according to the training result of each round, obtain an adjusted CNN neural network model, continue to train the adjusted CNN neural network model, and stop training when the training error is less than or equal to the training error threshold or the current number of iterations is greater than or equal to the maximum number of iterations, to obtain a trained CNN neural network model;

[0044] The above steps ensure that the model stops training after reaching the preset error threshold or the number of iterations, avoiding the problems of overfitting or underfitting; the trained CNN neural network model has higher generalization ability and prediction accuracy, can more reliably compensate and calibrate real-time water flow parameter data, and further enhances the data analysis and decision support ability of the fire extinguishing system.

[0045] Preferably, the S35 comprises the following steps:

[0046] S351, set a test accuracy threshold, input the historical environment and monitoring parameter data test matrix into the trained CNN neural network model for testing, obtain a test output result, compare the test output result with the historical actual water flow data test matrix, and obtain a test accuracy;

[0047] S352, when the test accuracy is greater than or equal to the test accuracy threshold, the trained CNN neural network model is used as a final CNN neural network model;

[0048] S353, when the test accuracy is less than the test accuracy threshold, the trained CNN neural network model is optimized by combining a grid search and a cross-validation algorithm, and after optimization, a final CNN neural network model is obtained;

[0049] The above steps can objectively evaluate the performance of the model by comparing the test output result with the historical actual data; if the test accuracy reaches or exceeds the threshold, the model is identified as qualified and can be directly used for practical application; if the threshold is not reached, the model is further optimized by using the grid search and cross-validation algorithm until the accuracy requirement is met; the prediction accuracy of the model is improved, ensuring the efficiency and accuracy of the CNN neural network model in real-time water flow parameter data compensation and calibration.

[0050] Preferably, the S353 combines the grid search binding with the cross-validation algorithm to optimize the trained CNN neural network model, and after optimization, the specific steps of obtaining the final CNN neural network model are as follows:

[0051] S3531, set the parameter search space as g , the maximum number of random searches is h , and the cross-validation fold number is set as X ;

[0052] S3532, initialize the optimal parameter combination , the best performance is initialized as v ; randomly select a parameter combination , construct a CNN neural network model according to the selected parameter combination, divide the training data into X folds, for each fold x , ; train the CNN neural network model using all training data except the x fold, and evaluate the model using the training data of the x fold to obtain a performance index set , calculate the average performance of all folds u , and the calculation formula is as follows,

[0053] ;

[0054] If the average performance u is greater than the best performance v , update the best performance to u , update the optimal parameter combination to , and update the training accuracy; otherwise, keep the original best performance, optimal parameter combination and training accuracy;

[0055] S3533, repeat S3532, when the maximum number of random searches g or the training accuracy is greater than or equal to the training accuracy threshold, stop iteration, obtain the optimal parameter combination, and use the optimal parameter combination as the parameter of the CNN neural network model to obtain the final CNN neural network model;

[0056] The above steps ensure that the optimal solution is found in a wide range of parameter combinations by setting the parameter search space and the maximum number of random searches, and by using the cross-validation fold number, while the best performance and parameter combination are maintained through iteration update.

[0057] Preferably, the back-end server is configured to calculate the water flow of each water gun and the remaining water volume of the vehicle-mounted water tank according to the compensated and calibrated water flow data, and adjust the water flow of each water gun according to the water flow of each water gun and the remaining water volume of the vehicle-mounted water tank. The specific steps are as follows:

[0058] S41, the back-end server establishes a fluid mechanics model according to the water flow parameter set, substitutes the water flow parameter data of each water gun in the compensated and corrected real-time water flow parameter data matrix into the fluid mechanics model, and obtains the water discharge of each water gun and the remaining water volume of the vehicle-mounted water tank;

[0059] S42, according to the water discharge of each water gun and the remaining water volume of the vehicle-mounted water tank, the back-end server sends an instruction to adjust the water discharge of the water gun;

[0060] The above steps realize intelligent control and optimization of the fire extinguishing process by establishing a fluid mechanics model, substituting the compensated and corrected real-time water flow parameter data into the calculation, and sending an instruction to adjust the water discharge of the water gun according to the calculation result of the back-end server. This design not only improves the utilization efficiency of water resources, but also enhances the accuracy and safety of fire extinguishing operations, effectively improves the overall performance and intelligent level of the fire extinguishing system, and provides more scientific and efficient fire extinguishing support for firefighters.

[0061] (Three) beneficial effects

[0062] The present application has the following beneficial effects:

[0063] The present application realizes accurate and real-time monitoring of the water discharge of the fire gun by introducing a flow monitoring device based on a wireless intelligent monitoring ring. Compared with the traditional estimation method relying on the experience and feeling of firefighters, the present application can significantly improve the accuracy of water volume evaluation and reduce errors, thereby improving the efficiency of fire extinguishing operations. Through real-time data collection and processing, fire commanders can more quickly and accurately grasp the remaining water volume of the vehicle-mounted water tank, providing a scientific basis for formulating and adjusting fire extinguishing strategies.

[0064] The monitoring device of the present application fully considers the actual environmental requirements of the fire scene. The circular ring-shaped shell has an impact protection layer and a shockproof structure to ensure that the device can still work stably in complex and variable fire scenes. The design of the heat dissipation holes enables the device to operate normally in high-temperature environments, while the quick connection device and the sealing ring ensure the convenient and firm connection of the device with the water gun and the water hose. These design features greatly enhance the adaptability and reliability of the device, providing strong technical support for firefighters.

[0065] The present application can realize accurate regulation of the water discharge of each water gun and real-time updating of the remaining water volume of the vehicle-mounted water tank. This not only helps to avoid waste of water resources, but also ensures that there is always enough water supply during the fire extinguishing process. In addition, based on accurate water volume data, fire commanders can more reasonably allocate and dispatch resources, optimize fire extinguishing decisions, improve fire extinguishing success rate, and minimize losses caused by fires.

[0066] Of course, implementing any product of the present application does not necessarily require achieving all the advantages described above at the same time. BRIEF DESCRIPTION OF DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, the drawings can also obtain drawings according to these drawings without creative labor.

[0068] Figure 1 It is a front view and side view structural schematic diagram of the bayonet type wireless intelligent monitoring ring in the flow monitoring device based on the wireless intelligent monitoring ring of the present application.

[0069] Figure 2 It is a flow process schematic diagram for realizing flow monitoring in the flow monitoring device based on the wireless intelligent monitoring ring of the present application.

[0070] Figure 3 It is a whole structure schematic diagram of the quick interface type wireless intelligent monitoring ring in the flow monitoring device based on the wireless intelligent monitoring ring of the present application.

[0071] In the figure:

[0072] 1, electronic bin; 2, circular ring shell; 3, measuring hole. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some 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 are within the scope of protection of the present application.

[0074] In the description of the present application, it should be understood that the terms "opening", "upper", "lower", "top", "middle", "inner" and the like indicate the orientation or positional relationship, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the indicated components or elements must have a particular orientation, be constructed and operated in a particular orientation, therefore it cannot be understood as a limitation of the present application.

[0075] Embodiment one:

[0076] Please refer to Figure 1 , Figure 2 , Figure 3The application discloses a flow monitoring device based on a wireless intelligent monitoring ring, which comprises a wireless intelligent monitoring ring and a back-end server; the caliber types of the wireless intelligent monitoring ring include 65 caliber, 80 caliber and the like; and the types of the wireless intelligent monitoring ring include a bayonet type and a quick interface type and the like.

[0077] The wireless intelligent monitoring ring comprises a circular ring-shaped shell 2 arranged between a water gun and a water hose, an electronic bin 1 and a measuring hole 3 arranged on the circular ring-shaped shell 2.

[0078] Quick connecting devices are arranged at both ends of the circular ring-shaped shell 2, and sealing rings are arranged at the connecting positions.

[0079] An impact-resistant protective layer is designed on the surface of the shell, and an anti-vibration structure is arranged on the inner wall of the shell.

[0080] The shell is designed with heat dissipation holes for ensuring that the device can still work normally in a high-temperature environment.

[0081] A sensor is arranged in the measuring hole 3 and used for measuring water flow parameters and environmental parameter data flowing through the water hose and the water gun in real time.

[0082] A microprocessor is arranged in the electronic bin 1 and comprises a signal acquisition unit, a data processing unit and a water quantity calculation unit.

[0083] The signal acquisition unit is connected with the sensor through a lead wire and used for acquiring water flow data and environmental data near the water gun.

[0084] The data processing unit is used for compensating and calibrating the water flow data according to the environmental data.

[0085] A data transmission unit is used for sending the water flow data compensated and calibrated to the back-end server.

[0086] The back-end server is used for calculating water discharge of each water gun and residual water quantity of a vehicle-mounted water tank according to the water flow data compensated and calibrated, and adjusting the water discharge of the water gun according to the water discharge of each water gun and the residual water quantity of the vehicle-mounted water tank.

[0087] The specific steps of the sensor arranged in the measuring hole 3 for measuring water flow parameters and environmental parameter data flowing through the water hose and the water gun in real time are as follows:

[0088] S11, setting a water flow parameter set S12, setting an environmental parameter set wherein a i the i-th water flow parameter in the water flow parameter set is represented as i the total number of water flow parameters is represented as n b i the i-th environmental parameter in the environmental parameter set is represented as i ​an environmental parameter, q representing the total number of environmental parameters; water flow parameters such as flow rate, flow volume, water pressure, etc., and environmental parameters such as temperature, humidity, etc.;

[0089] S12, arranging sensors in the measuring hole 3 of the wireless intelligent monitoring ring according to the water flow parameters and the environmental parameters;

[0090] The signal acquisition unit is connected with the sensor through a lead line, and is used for acquiring water flow data and environmental data near the water gun. The specific steps are as follows:

[0091] S21, collecting the number of real-time water guns to obtain the number of real-time water guns m , and m The wireless intelligent monitoring ring is connected with the water gun and the water hose through the quick connecting device at both ends of the circular ring-shaped shell 2. The quick connecting device is, for example, a buckle or a threaded interface.

[0092] The real-time water flow parameter data and the real-time environmental parameter data measured by the sensor through the water hose and the water gun are collected through the lead line to obtain a real-time monitoring water flow parameter data matrix A and a real-time environmental data parameter matrix B , as follows,

[0093] ;

[0094] ;

[0095] wherein, A in represents the real-time water flow parameter data matrix A , the i th parameter data of the real-time water flow of the n th water gun in the real-time water flow parameter data matrix B iq represents the i th parameter of the real-time environment of the q th water gun in the real-time environmental parameter data matrix;

[0096] The data processing unit is used for compensating and calibrating the water flow data according to the environmental data. The specific steps are as follows:

[0097] S31, collecting actual water flow parameter data under different historical environments and water flow parameter data measured by the wireless intelligent monitoring ring to obtain a historical environmental and monitoring parameter data matrix and a historical actual water flow parameter data matrix, and taking the historical actual water flow parameter data matrix as a label matrix of the historical environmental and monitoring parameter data matrix;

[0098] S32, constructing an initial CNN neural network model, and setting the initial weight of the initial CNN neural network model as c 1, the initial bias isd 1, the initial learning rate is e 1, set the training ratio and test ratio of the initial CNN neural network model as f 1, f 2;

[0099] S33, according to the f 1, f 2, the historical environment and monitoring parameter data matrix is divided into a historical environment and monitoring parameter data training matrix and a historical environment and monitoring parameter data test matrix; and the corresponding historical actual water flow parameter data matrix is divided into a historical actual water flow parameter data training matrix and a historical actual water flow parameter data test matrix;

[0100] S34, the historical environment and monitoring parameter data training matrix and the historical actual water flow parameter data training matrix are input into an initial CNN neural network model for training, and after the training is completed, a trained CNN neural network model is obtained;

[0101] The specific steps of the S34 are as follows:

[0102] S341, a training error threshold and a maximum number of iterations are set;

[0103] S342, the historical environment and monitoring parameter data training matrix and the historical actual water flow parameter data training matrix are input into the initial CNN neural network model, and the initial CNN neural network model is repeatedly trained;

[0104] S343, according to the training result of each round, the initial weight, the initial bias and the initial learning rate of the initial CNN neural network model are adjusted to obtain an adjusted CNN neural network model, and the adjusted CNN neural network model is continuously trained; when the training error is less than or equal to the training error threshold or the current number of iterations is greater than or equal to the maximum number of iterations, the training is stopped, and a trained CNN neural network model is obtained.

[0105] S35, the historical environment and monitoring parameter data test matrix and the historical actual water flow parameter data test matrix are input into the trained CNN neural network model for testing, and after the testing optimization is completed, a final CNN neural network model is obtained;

[0106] The specific steps of the S35 are as follows:

[0107] S351, a test accuracy threshold is set, the historical environment and monitoring parameter data test matrix is input into the trained CNN neural network model for testing, a test output result is obtained, the test output result is compared with the historical actual water flow data test matrix, and a test accuracy is obtained;

[0108] S352, when the test accuracy ≥ test accuracy threshold, the trained CNN neural network model is taken as a final CNN neural network model;

[0109] S353, when the test accuracy < test accuracy threshold, the trained CNN neural network model is optimized in combination with a grid search and a cross-validation algorithm, and after optimization, a final CNN neural network model is obtained;

[0110] The S353 has the following specific steps:

[0111] S3531, the parameter search space is set as g , the maximum random search number is h , and the cross-validation fold number is set as X ;

[0112] S3532, the optimal parameter combination is initialized , the best performance is initialized as v , a parameter combination is randomly selected , a CNN neural network model is constructed according to the selected parameter combination, the training data is divided into X folds, for each fold x , ; the CNN neural network model is trained using all training data except the x fold, the model is evaluated using the training data of the x fold, and a performance index set is obtained, the average performance of all folds is calculated as u , and the calculation formula is as follows,

[0113] ;

[0114] If the average performance u is greater than the best performance v , the best performance is updated as u , the optimal parameter combination is updated as , and the training accuracy is updated; otherwise, the original best performance, optimal parameter combination and training accuracy are kept;

[0115] S3533, S2532 is repeated, when the maximum random search number g or the training accuracy ≥ training accuracy threshold, the iteration is stopped, the optimal parameter combination is obtained, the optimal parameter combination is taken as the parameter of the CNN neural network model, and a final CNN neural network model is obtained;

[0116] S36, the real-time monitoring water flow parameter data matrix A and the real-time environmental parameter data matrix BThe data in the compensation correction real-time water flow parameter data matrix is substituted into the final CNN neural network model to obtain a compensation correction real-time water flow parameter data matrix C As follows,

[0117] ;

[0118] wherein, Cin represents the i th real-time water flow parameter data of the j th water gun after compensation correction in the compensation correction real-time water flow parameter data matrix i n

[0119] The back-end server is configured to calculate the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank according to the compensated and calibrated water flow data; and the specific steps of adjusting the water discharge of the water gun according to the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank are as follows:

[0120] S41, the back-end server establishes a fluid mechanics model according to the water flow parameter set, substitutes the water flow parameter data of each water gun in the compensation correction real-time water flow parameter data matrix into the fluid mechanics model, and obtains the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank;

[0121] S42, according to the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank, the back-end server sends an instruction to adjust the water discharge of the water gun.

[0122] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the invention. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0123] The preferred embodiments of the above disclosed invention are only used to help explain the invention. The preferred embodiments do not describe all the details and limit the invention to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can well understand and utilize the invention.​​

Claims

1. A flow monitoring device based on a wireless smart monitoring ring, characterized by, The application relates to a wireless intelligent monitoring ring and a backend server. The wireless intelligent monitoring ring comprises a circular shell arranged between a water gun and a water hose, an electronic bin and a measuring hole are arranged on the circular shell; Quick connecting devices are arranged at two ends of the circular shell, and sealing rings are arranged at connecting positions; An impact-resistant protective layer is arranged on the surface of the shell, and a shockproof structure is arranged on the inner wall of the shell; The shell is provided with a heat dissipation hole; A sensor is arranged in the measuring hole and used for measuring water flow parameters and environmental parameter data flowing through the water hose and the water gun in real time; A microprocessor is arranged in the electronic bin and comprises a signal acquisition unit, a data processing unit and a water quantity calculation unit; The signal acquisition unit is connected with the sensor through a lead wire and used for acquiring water flow data and environmental data near the water gun; The data processing unit is used for compensating and calibrating the water flow data according to the environmental data; A data transmission unit is used for sending the compensated and calibrated water flow data to the backend server, The backend server is used for calculating water discharge of each water gun and residual water quantity of a vehicle-mounted water tank according to the compensated and calibrated water flow data, and adjusting the water discharge of the water gun according to the water discharge of each water gun and the residual water quantity of the vehicle-mounted water tank. The specific steps that the sensor is arranged in the measuring hole and used for measuring water flow parameters and environmental parameter data flowing through the water hose and the water gun in real time are as follows: S12, arranging the sensor in the measuring hole of the wireless intelligent monitoring ring according to parameters in the water flow parameter set and the environmental parameter set; S11, set a water flow parameter set , set an environmental parameter set wherein represents an i-th water flow parameter in the water flow parameter set, n represents a total number of water flow parameters, represents an i-th environmental parameter in the environmental parameter set, q represents a total number of environmental parameters; The specific steps that the signal acquisition unit is connected with the sensor through a lead wire and used for acquiring water flow data and environmental data near the water gun are as follows: S21, collecting the number of real-time water guns to obtain a real-time water gun number m, and connecting the water gun and the water hose through the quick connecting devices at two ends of the circular shell of the m wireless intelligent monitoring rings; The specific steps that the data processing unit is used for compensating and calibrating the water flow data according to the environmental data are as follows: The real-time flow parameter data measured by the lead wire collection sensor flowing through the water hose and the water gun and the real-time environmental parameter data are obtained to obtain a real-time monitoring flow parameter data matrix A and a real-time environmental data parameter matrix B, as follows, wherein, represents the nth parameter data of the real-time water flow of the ith water gun in the real-time water flow parameter data matrix A, represents the qth parameter of the real-time environment of the ith water gun in the real-time environment parameter data matrix. S31, collecting actual water flow parameter data under different historical environments and water flow parameter data measured by the wireless intelligent monitoring ring to obtain a historical environmental and monitoring parameter data matrix and a historical actual water flow parameter data matrix, and taking the historical actual water flow parameter data matrix as a label matrix of the historical environmental and monitoring parameter data matrix; S34, inputting the historical environmental and monitoring parameter data training matrix and the historical actual water flow parameter data training matrix into an initial CNN neural network model for training, obtaining a trained CNN neural network model after the training is completed; S32, construct an initial CNN neural network model, set the initial weight of the initial CNN neural network model as , the initial bias as , and the initial learning rate as ; set the training proportion and the test proportion of the initial CNN neural network model as ; S33、according to the The historical environment and monitoring parameter data matrix is divided into a historical environment and monitoring parameter data training matrix and a historical environment and monitoring parameter data test matrix, and the historical actual water flow parameter data matrix is correspondingly divided into a historical actual water flow parameter data training matrix and a historical actual water flow parameter data test matrix. S35, inputting the historical environmental and monitoring parameter data test matrix and the historical actual water flow parameter data test matrix into the trained CNN neural network model for testing, and obtaining a final CNN neural network model after the testing optimization is completed; The specific steps that the backend server is used for calculating water discharge of each water gun and residual water quantity of a vehicle-mounted water tank according to the compensated and calibrated water flow data, and adjusting the water discharge of the water gun according to the water discharge of each water gun and the residual water quantity of the vehicle-mounted water tank are as follows: S36, substituting the data in the real-time monitoring water flow parameter data matrix A and the real-time environmental parameter data matrix B into the final CNN neural network model to obtain a compensated and corrected real-time water flow parameter data matrix C, as follows, ; wherein, represents the n-th real-time water flow parameter data of the i-th water gun in the compensated corrected real-time water flow parameter data matrix. ​ S41, the back-end server establishes a fluid mechanics model according to the water flow parameter set, substitutes the water flow parameter data of each water gun in the compensation correction real-time water flow parameter data matrix into the fluid mechanics model, and obtains the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank; S42, according to the water discharge of each water gun and the residual water volume of the vehicle-mounted water tank, the back-end server sends an instruction to adjust the water discharge of the water gun.

2. The flow monitoring device based on wireless smart monitoring ring of claim 1, wherein, The specific steps of S34 are as follows: S341, set a training error threshold and a maximum number of iterations; S342, input the historical environment and monitoring parameter data training matrix and the historical actual water flow parameter data training matrix into the initial CNN neural network model, and repeatedly train the initial CNN neural network model; S343, adjust the initial weight, initial bias and initial learning rate of the initial CNN neural network model according to the training result of each round, obtain the adjusted CNN neural network model, continue to train the adjusted CNN neural network model, and stop training when the training error is less than or equal to the training error threshold or the current iteration number is greater than or equal to the maximum number of iterations, to obtain the trained CNN neural network model.

3. The flow monitoring device based on wireless smart monitoring ring of claim 1, wherein, The specific steps of S35 are as follows: S351, set a test accuracy threshold, input the historical environment and monitoring parameter data test matrix into the trained CNN neural network model for testing, obtain the test output result, compare the test output result with the historical actual water flow data test matrix, and obtain the test accuracy; S352, when the test accuracy is greater than or equal to the test accuracy threshold, the trained CNN neural network model is used as the final CNN neural network model; S353, when the test accuracy is less than the test accuracy threshold, the trained CNN neural network model is optimized by combining the grid search and cross-validation algorithm, and the final CNN neural network model is obtained after optimization.

4. The flow monitoring device based on wireless smart monitoring ring of claim 3, wherein, The specific steps of S353, in which the trained CNN neural network model is optimized by combining the grid search and cross-validation algorithm, and the final CNN neural network model is obtained after optimization, are as follows: S3531, set the parameter search space as g, the maximum random search number as h, and the cross-validation fold number as X; S3532、initialize the optimal parameter combination , initialize the best performance v; randomly select a parameter combination , construct a CNN neural network model according to the selected parameter combination, divide the training data into X folds, for each fold x, ; train the CNN neural network model using all training data except the xth fold, evaluate the model using the training data of the xth fold, and obtain a performance indicator set , calculate the average of the performance indicators of all folds to obtain the average performance u, the calculation formula is as follows, ; If the average performance u > the best performance v, then the best performance is updated to u, and the optimal parameter combination is updated to and the training accuracy is updated; otherwise, the original best performance, the optimal parameter combination, and the training accuracy are kept. S3533, repeat S3532, and stop iteration when the maximum random search number g is reached or the training accuracy is greater than or equal to the training accuracy threshold, to obtain the optimal parameter combination, use the optimal parameter combination as the parameter of the CNN neural network model, and obtain the final CNN neural network model.

Images