Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a remote meter reading method, a remote meter reading system and electronic equipment for a water meter, which are used for solving the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme that the remote meter reading method of the water meter comprises the following steps:
Step one, acquiring water meter monitoring data of n intelligent water meters according to an electric signal, and generating water meter monitoring coefficients based on the water meter monitoring data, wherein n is an integer greater than zero, and the water meter monitoring data comprises actual measurement fluctuation data and preset standard fluctuation data;
Extracting a water meter monitoring coefficient time sequence set of the nth intelligent water meter, inputting historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set into a pre-constructed first deep learning model, and predicting the water meter monitoring coefficients at the future time T;
step three, presetting a gradient threshold, and comparing future T-moment water meter monitoring coefficients corresponding to the n intelligent water meters with the preset gradient threshold to obtain L abnormal intelligent water meters, wherein L is E n;
step four, obtaining abnormal characteristic data of the L-th abnormal intelligent water meter, determining fault type data of the abnormal intelligent water meter according to the abnormal characteristic data, and determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault analysis data comprises E abnormal reasons and a standard abnormal vibration FFT graph related to the E abnormal reasons;
Fifthly, acquiring an actually measured abnormal vibration FFT (fast Fourier transform) graph of the L-th abnormal intelligent water meter in a preset time interval according to the position of the fault element, comparing the actually measured abnormal vibration FFT graph with a standard abnormal vibration FFT graph, and determining the abnormal reason of the abnormal intelligent water meter according to a comparison result.
Further, generating a water meter monitoring coefficient based on the water meter monitoring data includes:
obtaining actual measurement fluctuation data of an nth intelligent water meter in a preset time interval based on water meter monitoring data, wherein the actual measurement fluctuation data comprises a fluctuation water volume value, a fluctuation temperature value and a fluctuation pressure value,
Extracting preset standard fluctuation data from a system database, wherein the preset standard fluctuation data comprise a preset standard water content value, a preset standard temperature value and a preset standard pressure value;
The preset standard fluctuation data is an average value calculated by historical actual measurement fluctuation data in a system database;
Carrying out formula calculation on the fluctuation water volume value and a preset standard water volume value, and calculating a water yield evaluation coefficient, wherein the calculation formula of the water yield evaluation coefficient is as follows: Wherein Lpn represents the water quantity evaluation coefficient of the nth intelligent water meter, lyn represents the fluctuation water quantity value of the nth intelligent water meter, ly'n represents the preset standard water quantity value of the nth intelligent water meter, and alphan represents the correction constant of the water quantity evaluation coefficient of the nth intelligent water meter;
carrying out formula calculation on the fluctuation temperature value and a preset standard temperature value, and calculating a water Wen Pinggu coefficient, wherein the calculation formula of the water temperature evaluation coefficient is as follows: Wherein Wpn represents the water temperature evaluation coefficient of the nth intelligent water meter, wdn represents the fluctuation temperature value of the nth intelligent water meter, wd'n represents the preset standard temperature value of the nth intelligent water meter, and αn represents the correction constant of the water temperature evaluation coefficient of the nth intelligent water meter;
carrying out formulated calculation on the fluctuation pressure value and a preset standard pressure value, and calculating a water pressure evaluation coefficient, wherein the calculation formula of the water pressure evaluation coefficient is as follows: wherein Ypn represents the water pressure evaluation coefficient of the nth intelligent water meter, ysn represents the fluctuation pressure value of the nth intelligent water meter, ys'n represents the preset standard pressure value of the nth intelligent water meter, and gamman represents the correction constant of the water pressure evaluation coefficient of the nth intelligent water meter;
Carrying out dimensionless treatment on the water quantity evaluation coefficient, the water temperature evaluation coefficient and the water pressure evaluation coefficient, and carrying out associated calculation to obtain a water meter monitoring coefficient, wherein the calculation formula of the water meter monitoring coefficient is as follows:
Wherein SJn denotes a water meter monitoring coefficient of the nth intelligent water meter, μ1 denotes a weight factor of a water volume evaluation coefficient, μ2 denotes a weight factor of a water temperature evaluation coefficient, and μ3 denotes a weight factor of a water pressure evaluation coefficient.
Further, the method for obtaining the water meter monitoring coefficient time sequence set comprises the following steps:
extracting water meter monitoring coefficients in the history monitoring process of the nth intelligent water meter from a system database, marking the water meter monitoring coefficients as history water meter monitoring coefficients, constructing a water meter monitoring coefficient time sequence set by the extracted history water meter monitoring coefficients, wherein the water meter monitoring coefficient time sequence set comprises i history water meter monitoring coefficients, the time intervals acquired by the i history water meter monitoring coefficients are equal, the i history water meter monitoring coefficients correspond to an intelligent water meter reading period, and the intelligent water meter reading period is day, month or year.
Further, the pre-construction logic of the first deep learning model is that a time step S, a sliding step B and a sliding window length H are preset according to the actual experience of a water service manager; converting historical water meter monitoring coefficients in a water meter monitoring coefficient time sequence set into a plurality of training samples by using a sliding window method, taking the training samples as input of a first deep learning model, taking the water meter monitoring coefficients after a prediction time step S as output, taking the subsequent water meter monitoring coefficients of each training sample as a prediction target, taking the prediction accuracy as a training target, training the first deep learning model, and generating the first deep learning model for predicting the water meter monitoring coefficients at the future T moment according to the historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set, wherein the first deep learning model is an RNN neural network model.
Further, logic for obtaining L abnormal intelligent water meters, comprising:
Preset gradient thresholds Yz1 and Yz2, wherein Yz1>Yz2, comparing future time T water meter monitoring coefficient SJn with the preset gradient threshold;
if SJn>Yz1, the corresponding intelligent water meter is an abnormal intelligent water meter;
If Yz1≥SJn≥Yz2 is detected, judging that the corresponding intelligent water meter is a normal intelligent water meter;
If SJn<Yz2, the corresponding intelligent water meter is an abnormal intelligent water meter;
counting all abnormal intelligent water meters to obtain L abnormal intelligent water meters.
Further, obtaining abnormal characteristic data of the L-th abnormal intelligent water meter includes:
step 301, acquiring a water pressure data set of an abnormal intelligent water meter, and constructing an actual measurement water pressure trend graph by taking time in the water pressure data set as a horizontal axis and water pressure data in the water pressure data set as a vertical axis;
Step 302, dividing the water pressure trend graph in equal parts according to a preset time interval to obtain an actual measured water pressure line graph set, wherein the actual measured water pressure line graph set comprises G actual measured water pressure line graphs, and G is an integer greater than zero;
Step 303, extracting a G-th measured water pressure line graph in the measured water pressure line graph set, wherein the initial value of G epsilon G and G is 1;
Step 304, obtaining a corresponding standard water pressure interval of the intelligent water meter, extracting a standard water pressure line graph related to the corresponding standard water pressure interval, calculating the similarity between the measured water pressure line graph and the standard water pressure line graph, jumping to step 305 if the similarity between the measured water pressure line graph and the standard water pressure line graph is larger than or equal to a preset water pressure similarity threshold value, marking the measured water pressure line graph as an abnormal water pressure line graph if the similarity between the measured water pressure line graph and the standard water pressure line graph is smaller than a preset vibration similarity threshold value, and jumping to step 305;
step 305. Let g=g+1 and jump back to step 303;
step 306, repeating the steps 303-305 until g=g, ending the cycle, and obtaining a plurality of abnormal water pressure line graphs;
Step 307, extracting the similarity corresponding to each abnormal water pressure line graph, and carrying out Fourier transform on the abnormal water pressure line graph with the maximum similarity to obtain an abnormal water pressure frequency spectrum graph, wherein the Fourier transform is specifically one of fast Fourier transform and short-time Fourier transform;
acquiring abnormal characteristic data of the L-th abnormal intelligent water meter, and further comprising:
Step 401, acquiring a current data set of an abnormal intelligent water meter, and constructing a current trend graph by taking time in the current data set as a horizontal axis and taking current data in the current data set as a vertical axis;
Step 402, dividing the current trend graph in equal parts according to a preset time interval to obtain an actual measurement current line graph set, wherein the actual measurement current line graph set comprises D actual measurement current line graphs, and D is an integer greater than zero;
step 403, extracting a d-th measured current line graph in the measured current line graph set, wherein d is a positive integer greater than zero, and the initial value of d is 1;
Step 404, obtaining a corresponding standard current interval of the intelligent water meter, extracting a standard current line graph associated with the corresponding standard current interval, calculating the similarity between the actually measured current line graph and the standard current line graph, jumping to step 405 if the similarity between the actually measured current line graph and the standard current line graph is greater than or equal to a preset current similarity threshold value, marking the actually measured current line graph as an abnormal current line graph if the similarity between the actually measured current line graph and the standard current line graph is less than a preset vibration similarity threshold value, and jumping to step 405;
Step 405, let h=h+1, and jump back to step 403;
step 406, repeating the steps 403 to 405 until d=d, ending the cycle, and obtaining a plurality of abnormal current line diagrams;
Step 407, extracting the similarity corresponding to each abnormal current line graph, and carrying out Fourier transform on the abnormal current line graph with the maximum similarity to obtain an abnormal current spectrum graph, wherein the Fourier transform is specifically one of fast Fourier transform and short-time Fourier transform.
Further, determining fault type data of the abnormal intelligent water meter includes:
Acquiring an abnormal water pressure spectrogram and an abnormal current spectrogram of an abnormal intelligent water meter;
Inputting the abnormal water pressure spectrogram and the abnormal current spectrogram into a fault identification model to determine fault type data of the abnormal intelligent water meter;
the generation logic of the fault recognition model comprises the steps of obtaining historical fault data which are prestored in a system database and used for training the fault recognition model, wherein the historical fault data comprise an abnormal water pressure spectrogram, an abnormal current spectrogram and fault types, dividing the historical fault data used for training the fault recognition model into a type training set and a type testing set, constructing a regression network model, taking the abnormal water pressure spectrogram and the abnormal current spectrogram in the type training set as input of the regression network model, taking the fault types in the type training set as output of the regression network model, training the regression network model to obtain an initial regression network model, carrying out model test on the initial regression network model by utilizing the type testing set, and screening the corresponding initial regression network model which is greater than or equal to preset test accuracy as the fault recognition model, wherein the regression network model comprises a random regression algorithm model, a support vector machine regression algorithm model and a neural network algorithm model.
Further, comparing the measured abnormal vibration FFT graph with the standard abnormal vibration FFT graph, and determining the abnormal reason of the abnormal intelligent water meter according to the comparison result, wherein the method comprises the following steps:
acquiring a vibration frequency fluctuation diagram of the intelligent water meter in a preset time interval, and performing Fourier transformation on the vibration frequency fluctuation diagram to obtain an actually measured abnormal vibration FFT diagram, wherein the vibration frequency fluctuation diagram is generated by r vibration frequencies acquired in real time in the preset time interval by a vibration sensor, and r is an integer larger than 1;
Vectorizing the actually-measured abnormal vibration FFT graph and the standard abnormal vibration FFT graph, and calculating the image similarity of the vectorized actually-measured abnormal vibration FFT graph and the standard abnormal vibration FFT graph according to a cosine similarity model;
judging whether the image similarity is larger than a preset image similarity threshold, if so, generating a comparison result, and extracting an abnormal reason corresponding to a standard abnormal vibration FFT graph as an abnormal reason of the abnormal intelligent water meter according to the comparison result.
In a second aspect, the present invention provides a remote meter reading system for a water meter, for implementing the above remote meter reading method for a water meter, including:
The data generation module is used for acquiring water meter monitoring data of n intelligent water meters according to the electric signals and generating water meter monitoring coefficients based on the water meter monitoring data, wherein n is an integer greater than zero, and the water meter monitoring data comprises actual measurement fluctuation data and preset standard fluctuation data;
The data prediction module is used for extracting a water meter monitoring coefficient time sequence set of the nth intelligent water meter, inputting a history water meter monitoring coefficient in the water meter monitoring coefficient time sequence set into a pre-constructed first deep learning model, and predicting a water meter monitoring coefficient at a future time T;
The data comparison module is used for presetting a gradient threshold value and comparing future T-moment water meter monitoring coefficients corresponding to the n intelligent water meters with the preset gradient threshold value to obtain L abnormal intelligent water meters, wherein L is E n;
The system comprises a data analysis module, a fault analysis module and a fault analysis module, wherein the data analysis module is used for acquiring abnormal characteristic data of an L-th abnormal intelligent water meter, determining fault type data of the abnormal intelligent water meter according to the abnormal characteristic data, and determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault analysis data comprises E abnormal reasons and a standard abnormal vibration FFT (fast Fourier transform) graph associated with the E abnormal reasons;
The data judging module is used for acquiring an actually measured abnormal vibration FFT (fast Fourier transform) graph of the L-th abnormal intelligent water meter in a preset time interval according to the position of the fault element, comparing the actually measured abnormal vibration FFT graph with a standard abnormal vibration FFT graph, and determining the abnormal reason of the abnormal intelligent water meter according to the comparison result.
In a third aspect, the invention provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program for the processor to call;
the processor executes the remote meter reading method of the water meter by calling the computer program stored in the memory.
The invention has the technical effects and advantages that:
the method comprises the steps of acquiring water meter monitoring data of n intelligent water meters according to electric signals, generating water meter monitoring coefficients based on the water meter monitoring data, extracting a water meter monitoring coefficient time sequence set of an nth intelligent water meter, inputting historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set into a pre-built first deep learning model, predicting water meter monitoring coefficients at a future time T, comparing the water meter monitoring coefficients corresponding to the n intelligent water meters with a preset gradient threshold to obtain L abnormal intelligent water meters, acquiring abnormal characteristic data of the L abnormal intelligent water meters, determining fault type data of the abnormal intelligent water meters according to the abnormal characteristic data, determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault type data comprises a fault type and a fault element position, acquiring an actually measured abnormal vibration FFT graph of the L abnormal intelligent water meters in a preset time interval according to the fault element position, comparing the abnormal vibration FFT graph with a standard abnormal vibration FFT graph, and determining abnormal reasons of the abnormal intelligent water meters according to comparison results, and being beneficial to the water meters sent by intelligent water meters, and being convenient to quickly and capable of quickly and rapidly replacing the intelligent water meters according to the reasons of the fault monitoring data.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.
It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and a similar second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Example 1
Referring to fig. 1, the disclosure of the present embodiment provides a method for remote reading of a water meter, including:
Step one, acquiring water meter monitoring data of n intelligent water meters according to an electric signal, and generating water meter monitoring coefficients based on the water meter monitoring data, wherein n is an integer greater than zero, and the water meter monitoring data comprises actual measurement fluctuation data and preset standard fluctuation data;
It should be noted that, the N intelligent water meters are preset in the target city according to the urban water pipe topology network, the N intelligent water meters are each provided with a remote meter reading device, the remote meter reading devices receive the electric signals of the remote server and collect the water meter monitoring data, the water meter monitoring data are stored in the system database, the remote meter reading devices are provided with a plurality of sensors, the sensors include but are not limited to flow sensors, temperature sensors, vibration sensors or pressure sensors, etc., the types of the remote meter reading devices are determined by water service management personnel according to the conditions, and the types of the remote meter reading devices are not excessively limited.
In a preferred embodiment, generating the water meter monitoring coefficients based on the water meter monitoring data includes:
101. Obtaining actual measurement fluctuation data of an nth intelligent water meter in a preset time interval based on water meter monitoring data, wherein the actual measurement fluctuation data comprises a fluctuation water volume value, a fluctuation temperature value and a fluctuation pressure value,
102. Extracting preset standard fluctuation data from a system database, wherein the preset standard fluctuation data comprise a preset standard water content value, a preset standard temperature value and a preset standard pressure value;
The preset standard fluctuation data is an average value calculated by historical actual measurement fluctuation data in a system database;
103. Carrying out formula calculation on the fluctuation water volume value and a preset standard water volume value, and calculating a water yield evaluation coefficient, wherein the calculation formula of the water yield evaluation coefficient is as follows: wherein Lpn represents the water quantity evaluation coefficient of the nth intelligent water meter, lyn represents the fluctuation water quantity value of the nth intelligent water meter, ly'n represents the preset standard water quantity value of the nth intelligent water meter, and alphan represents the correction constant of the water quantity evaluation coefficient of the nth intelligent water meter;
104. Carrying out formula calculation on the fluctuation temperature value and a preset standard temperature value, and calculating a water Wen Pinggu coefficient, wherein the calculation formula of the water temperature evaluation coefficient is as follows: Wherein Wpn represents the water temperature evaluation coefficient of the nth intelligent water meter, wdn represents the fluctuation temperature value of the nth intelligent water meter, wd'n represents the preset standard temperature value of the nth intelligent water meter, and αn represents the correction constant of the water temperature evaluation coefficient of the nth intelligent water meter;
105. Carrying out formulated calculation on the fluctuation pressure value and a preset standard pressure value, and calculating a water pressure evaluation coefficient, wherein the calculation formula of the water pressure evaluation coefficient is as follows: Wherein Wpn represents the water pressure evaluation coefficient of the nth intelligent water meter, ysn represents the fluctuation pressure value of the nth intelligent water meter, wd'n represents the preset standard pressure value of the nth intelligent water meter, and γn represents the correction constant of the water pressure evaluation coefficient of the nth intelligent water meter;
106. Carrying out dimensionless treatment on the water quantity evaluation coefficient, the water temperature evaluation coefficient and the water pressure evaluation coefficient, and carrying out associated calculation to obtain a water meter monitoring coefficient, wherein the calculation formula of the water meter monitoring coefficient is as follows:
wherein SJn represents the water meter monitoring coefficient of the nth intelligent water meter, μ1 represents the weight factor of the water volume evaluation coefficient, μ2 represents the weight factor of the water temperature evaluation coefficient, μ3 represents the weight factor of the water pressure evaluation coefficient;
Extracting a water meter monitoring coefficient time sequence set of the nth intelligent water meter, inputting historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set into a pre-constructed first deep learning model, and predicting the water meter monitoring coefficients at the future time T;
It should be noted that the method for obtaining the time series set of the water meter monitoring coefficients includes:
Extracting water meter monitoring coefficients in the history monitoring process of the nth intelligent water meter from a system database, marking the water meter monitoring coefficients as history water meter monitoring coefficients, constructing a water meter monitoring coefficient time sequence set by the extracted history water meter monitoring coefficients, wherein the water meter monitoring coefficient time sequence set comprises i history water meter monitoring coefficients, the time intervals acquired by the i history water meter monitoring coefficients are equal, and the i history water meter monitoring coefficients correspond to one intelligent water meter reading period;
the meter reading period of the intelligent water meter can be day, month or year, or can be set by a water service manager according to actual experience, and the method is not particularly limited.
The pre-construction logic of the first deep learning model comprises the steps of presetting a time step S, a sliding step B and a sliding window length H according to actual experience of water management staff, converting historical water meter monitoring coefficients in a water meter monitoring coefficient time sequence set into a plurality of training samples by using a sliding window method, taking the training samples as input of the first deep learning model, taking the water meter monitoring coefficients after the prediction time step S as output, taking the follow-up water meter monitoring coefficients of each training sample as a prediction target, taking the prediction accuracy rate as a training target, training the first deep learning model, generating a first deep learning model for predicting future water meter monitoring coefficients at the moment T according to the historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set, wherein the first deep learning model can be an RNN neural network model, predicting the future water meter monitoring coefficients of n intelligent water meters by using the water meter monitoring coefficient time sequence set, and accordingly realizing the prediction of the water meter monitoring coefficients to be predicted on the aspect of intelligent meter reading based on the future water meter reading conditions;
The method comprises the steps of dividing a time sequence set of water meter monitoring coefficients into A sliding windows with the same size, taking the water meter monitoring coefficients in each window as one sample, taking the water meter monitoring coefficients of the window at the future time T as digital labels, and the method comprises the steps that one sample corresponds to one digital label, one sample and the corresponding digital label form training data, and a plurality of groups of training data form a training set;
By way of further illustration, assuming that the time series set of meter monitor coefficients a contains 10 sets of historical meter monitor coefficients, a= { a1,A2,A3,A4,A5,...,A10},Aa is the a-th set of historical meter monitor coefficients, a sliding window is used to construct a plurality of training samples, a prediction time step is set to be 1, a length H of the sliding window is set to be 5, and a sliding step B is set to be 1, each training sample is generated to contain consecutive 5 historical meter monitor coefficients, and a next meter monitor coefficient of the consecutive 5 historical meter monitor coefficients is taken as a prediction target, for example:
{ A1,A2,A3,A4,A5 } is used as training data, and the prediction target corresponding to the training data is A6;
{ A2,A3,A4,A5,A6 } is used as training data, and the prediction target corresponding to the training data is A7;
Similarly, a first deep learning model is used for training the water meter monitoring coefficient for predicting the future moment.
Step three, presetting a gradient threshold, and comparing future T-moment water meter monitoring coefficients corresponding to the n intelligent water meters with the preset gradient threshold to obtain L abnormal intelligent water meters, wherein L is E n;
logic for obtaining L abnormal intelligent water meters, comprising:
201. Preset gradient thresholds Yz1 and Yz2, wherein Yz1>Yz2, comparing the future time T water meter monitoring coefficient SJn with the preset gradient threshold;
202. If SJn>Yz1, the corresponding intelligent water meter is an abnormal intelligent water meter;
If Yz1≥SJn≥Yz2 is detected, judging that the corresponding intelligent water meter is a normal intelligent water meter;
If SJn<Yz2, the corresponding intelligent water meter is an abnormal intelligent water meter;
203. counting all abnormal intelligent water meters to obtain L abnormal intelligent water meters;
Step four, obtaining abnormal characteristic data of the L-th abnormal intelligent water meter, determining fault type data of the abnormal intelligent water meter according to the abnormal characteristic data, and determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault analysis data comprises E abnormal reasons and a standard abnormal vibration FFT (fast Fourier transform) chart associated with the E abnormal reasons;
specifically, the abnormal characteristic data include an abnormal water pressure spectrogram and an abnormal current spectrogram;
in one specific embodiment, acquiring abnormal characteristic data of the L-th abnormal intelligent water meter includes:
301. The method comprises the steps of acquiring a water pressure data set of an abnormal intelligent water meter, and constructing an actual measurement water pressure trend graph by taking time in the water pressure data set as a horizontal axis and water pressure data in the water pressure data set as a vertical axis;
302. Dividing the water pressure trend graph in equal parts according to a preset time interval to obtain an actual measured water pressure line graph set, wherein the actual measured water pressure line graph set comprises G actual measured water pressure line graphs, and G is an integer greater than zero;
303. Extracting a G-th measured water pressure line graph in the measured water pressure line graph set, wherein G epsilon G, and the initial value of G is 1;
304. Acquiring a corresponding standard water pressure interval of the intelligent water meter, extracting a standard water pressure line graph related to the corresponding standard water pressure interval, calculating the similarity between the measured water pressure line graph and the standard water pressure line graph, and jumping to the step 305 if the similarity between the measured water pressure line graph and the standard water pressure line graph is greater than or equal to a preset water pressure similarity threshold value;
305. let g=g+1 and jump back to step 303;
306. Repeating the steps 303-305 until g=g, ending the cycle, and obtaining a plurality of abnormal water pressure line diagrams;
307. Extracting the similarity corresponding to each abnormal water pressure line graph, and carrying out Fourier transform on the abnormal water pressure line graph with the maximum similarity to obtain an abnormal water pressure frequency spectrum graph;
in one specific embodiment, acquiring the abnormal characteristic data of the L-th abnormal intelligent water meter further comprises:
401. The method comprises the steps of obtaining a current data set of an abnormal intelligent water meter, and constructing a current trend graph by taking time in the current data set as a horizontal axis and current data in the current data set as a vertical axis;
402. Dividing the current trend graph in equal parts according to a preset time interval to obtain an actual measurement current line graph set, wherein the actual measurement current line graph set comprises D actual measurement current line graphs, and D is an integer greater than zero;
403. Extracting a d-th measured current line diagram in the measured current line diagram set, wherein d is a positive integer greater than zero, and the initial value of d is 1;
404. Acquiring a corresponding standard current interval of the intelligent water meter, extracting a standard current line graph associated with the corresponding standard current interval, calculating the similarity between the actually measured current line graph and the standard current line graph, and jumping to the step 405 if the similarity between the actually measured current line graph and the standard current line graph is greater than or equal to a preset current similarity threshold value;
the system database is pre-stored with a plurality of current intervals, each current interval is associated with a standard current line diagram, and the standard current line diagram reflects the normal current representation of the intelligent water meter under the condition of no anomaly;
405. let h=h+1 and jump back to step 403;
406. Repeating the steps 403-405 until d=d, ending the cycle, and obtaining a plurality of abnormal current line diagrams;
407. Extracting the similarity corresponding to each abnormal current line graph, and carrying out Fourier transformation on the abnormal current line graph with the maximum similarity to obtain an abnormal current spectrogram;
the fourier transform is specifically one of a fast fourier transform or a short-time fourier transform;
in an implementation, determining fault type data for an abnormal intelligent water meter includes:
Acquiring an abnormal water pressure spectrogram and an abnormal current spectrogram of an abnormal intelligent water meter;
Inputting the abnormal water pressure spectrogram and the abnormal current spectrogram into a fault identification model to determine fault type data of the abnormal intelligent water meter;
The generation logic of the fault recognition model comprises the steps of acquiring historical fault data which is pre-stored in a system database and is used for training the fault recognition model, wherein the historical fault data comprises an abnormal water pressure spectrogram, an abnormal current spectrogram and a fault type; dividing historical fault data for training a fault recognition model into a type training set and a type testing set, constructing a regression network model, taking an abnormal water pressure spectrogram and an abnormal current spectrogram in the type training set as inputs of the regression network model, taking fault types in the type training set as outputs of the regression network model, and training the regression network model to obtain an initial regression network model;
Wherein the regression network model comprises, but is not limited to, a random forest regression algorithm model, a support vector machine regression algorithm model, a neural network algorithm model and the like;
It should be noted that the fault type includes u fault element names, such as a mechanical element (a gear, a mechanical transmission device or a metering pointer), a battery element (a battery or an external power source), a sensor (a flow sensor, a temperature sensor or a water pressure sensor), and the like, and the fault element position is determined according to the fault element names, and specifically, the installation position of the corresponding fault element can be obtained according to the design drawing of the intelligent water meter.
Step five, acquiring an actually measured abnormal vibration FFT (fast Fourier transform) diagram of the L-th abnormal intelligent water meter in a preset time interval according to the position of the fault element, comparing the actually measured abnormal vibration FFT diagram with a standard abnormal vibration FFT diagram, and determining an abnormal reason of the abnormal intelligent water meter according to a comparison result;
Specifically, comparing the actually measured abnormal vibration FFT graph with the standard abnormal vibration FFT graph, and determining an abnormal reason of the abnormal intelligent water meter according to a comparison result, wherein the method comprises the following steps:
acquiring a vibration frequency fluctuation diagram of the intelligent water meter in a preset time interval, and performing Fourier transformation on the vibration frequency fluctuation diagram to obtain an actually measured abnormal vibration FFT diagram, wherein the vibration frequency fluctuation diagram is generated by r vibration frequencies acquired in real time in the preset time interval by a vibration sensor, and r is an integer larger than 1;
Vectorizing the actually-measured abnormal vibration FFT graph and the standard abnormal vibration FFT graph, and calculating the image similarity of the vectorized actually-measured abnormal vibration FFT graph and the standard abnormal vibration FFT graph according to a cosine similarity model;
It should be noted that the purpose of vectorizing the measured abnormal vibration FFT graph with the standard abnormal vibration FFT graph is to convert the complex vibration spectrum data into a form that is easier to process and analyze, so as to better identify and understand the patterns and features in the vibration data.
Judging whether the image similarity is larger than a preset image similarity threshold value, if so, generating a comparison result, and extracting an abnormal reason corresponding to a standard abnormal vibration FFT graph as an abnormal reason of the abnormal intelligent water meter according to the comparison result;
If the comparison result which is larger than the preset image similarity threshold value does not exist, displaying a null value;
optionally, the method further comprises:
judging whether the comparison result has a null value or not, if so, recording that the corresponding actually measured abnormal vibration FFT image is a non-prestored image;
updating the corresponding relation between the second standard abnormal vibration FFT graph and the fault analysis data according to the non-prestored image;
The method comprises the steps of acquiring corresponding position data of the abnormal intelligent water meter in the urban water pipe topological structure network according to a comparison result or a non-prestored image, sending fault early warning to a system server, facilitating water service management personnel to reach the corresponding position of the abnormal intelligent water meter in time, and rapidly maintaining or replacing the abnormal intelligent water meter according to the fault cause of the abnormal intelligent water meter, so that loss caused by the abnormal intelligent water meter is reduced.
The remote meter reading device is in remote communication connection with the intelligent water meter.
The method comprises the steps of acquiring water meter monitoring data of n intelligent water meters according to electric signals, generating water meter monitoring coefficients based on the water meter monitoring data, extracting a water meter monitoring coefficient time sequence set of an nth intelligent water meter, inputting historical water meter monitoring coefficients in the water meter monitoring coefficient time sequence set into a pre-built first deep learning model, predicting water meter monitoring coefficients at a future time T, comparing the water meter monitoring coefficients at the future time T corresponding to the n intelligent water meters with a preset gradient threshold to obtain L abnormal intelligent water meters, acquiring abnormal characteristic data of the L abnormal intelligent water meters, determining fault type data of the abnormal intelligent water meters according to the abnormal characteristic data, determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault analysis data comprises E abnormal reasons and standard abnormal vibration FFT graphs associated with the E abnormal reasons, acquiring actual measurement abnormal vibration FFT graphs of the L abnormal intelligent water meters in a preset time interval according to the fault element positions, comparing the actual measurement abnormal vibration FFT graphs with the FFT graphs to the L abnormal intelligent water meters, determining that the intelligent water meters can be quickly replaced according to the fault type data and the intelligent water meter monitoring causes, and the fault management of the intelligent water meters can be quickly replaced according to the fault causes, and the fault causes can be quickly replaced by the intelligent water meters, and the personnel can quickly find the reasons of the abnormal intelligent water meters can be easily and the fault-conveniently and the reasons can be quickly replaced.
The formula related in the above description is a formula for removing dimension and taking numerical calculation, and is obtained by acquiring a large amount of data and performing software simulation to obtain the closest real situation, wherein the weight factors in the formula and each preset threshold value in the analysis process are set by a person skilled in the art according to the real situation or are obtained by performing large amount of data simulation, the size of the weight factors is a specific numerical value obtained by quantizing each parameter, the subsequent comparison is convenient, the size of the weight factors depends on the number of sample data and the corresponding processing coefficient initially set by the person skilled in the art for each group of sample data, and the proportional relation between the parameters and the quantized numerical value is not influenced.
Example 2
Referring to fig. 2, the present embodiment provides a remote meter reading system for a water meter, including:
The data generation module is used for acquiring water meter monitoring data of n intelligent water meters according to the electric signals and generating water meter monitoring coefficients based on the water meter monitoring data, wherein n is an integer greater than zero, and the water meter monitoring data comprises actual measurement fluctuation data and preset standard fluctuation data;
The data prediction module is used for extracting a water meter monitoring coefficient time sequence set of the nth intelligent water meter, inputting a history water meter monitoring coefficient in the water meter monitoring coefficient time sequence set into a pre-constructed first deep learning model, and predicting a water meter monitoring coefficient at a future time T;
The data comparison module is used for presetting a gradient threshold value and comparing future T-moment water meter monitoring coefficients corresponding to the n intelligent water meters with the preset gradient threshold value to obtain L abnormal intelligent water meters, wherein L is E n;
The system comprises a data analysis module, a fault analysis module and a fault analysis module, wherein the data analysis module is used for acquiring abnormal characteristic data of an L-th abnormal intelligent water meter, determining fault type data of the abnormal intelligent water meter according to the abnormal characteristic data, and determining fault analysis data corresponding to the fault type data based on a preset relation between the fault type data and the fault analysis data, wherein the fault analysis data comprises E abnormal reasons and a standard abnormal vibration FFT (fast Fourier transform) graph associated with the E abnormal reasons;
The data judging module is used for acquiring an actually measured abnormal vibration FFT (fast Fourier transform) graph of the L-th abnormal intelligent water meter in a preset time interval according to the position of the fault element, comparing the actually measured abnormal vibration FFT graph with a standard abnormal vibration FFT graph, and determining the abnormal reason of the abnormal intelligent water meter according to the comparison result.
Example 3
Referring to fig. 3, the present embodiment provides an electronic device, including a processor and a memory, where the memory stores a computer program for the processor to call;
The processor executes a remote meter reading method of the water meter of embodiment 1 by calling a computer program stored in the memory.
Example 4
Referring to fig. 4, the present embodiment provides a computer readable storage medium storing instructions that, when executed on a computer, cause the computer to perform a method for remote meter reading of a water meter of embodiment 1.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, from one website site, computer, server, or data center over a wired network. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely one, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Finally, the foregoing description of the preferred embodiment of the invention is provided for the purpose of illustration only, and is not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.