CN117786325B - Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge - Google Patents
Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge Download PDFInfo
- Publication number
- CN117786325B CN117786325B CN202410210577.9A CN202410210577A CN117786325B CN 117786325 B CN117786325 B CN 117786325B CN 202410210577 A CN202410210577 A CN 202410210577A CN 117786325 B CN117786325 B CN 117786325B
- Authority
- CN
- China
- Prior art keywords
- window
- data
- step length
- sequence
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 25
- 239000004429 Calibre Substances 0.000 title claims description 3
- 230000006855 networking Effects 0.000 title claims description 3
- 238000005070 sampling Methods 0.000 claims abstract description 70
- 230000003044 adaptive effect Effects 0.000 claims abstract description 44
- 230000007613 environmental effect Effects 0.000 claims abstract description 16
- 230000006978 adaptation Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000012937 correction Methods 0.000 claims description 11
- 238000012512 characterization method Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 description 13
- 230000000694 effects Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- VEMKTZHHVJILDY-UHFFFAOYSA-N resmethrin Chemical compound CC1(C)C(C=C(C)C)C1C(=O)OCC1=COC(CC=2C=CC=CC=2)=C1 VEMKTZHHVJILDY-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an intelligent early warning system for the ambient temperature of a large-caliber Internet of things water meter; obtaining an initial window step length and an initial sampling sequence according to the data change characteristics of the temperature data sequence; selecting an initial window step length according to the initial sampling sequence to obtain an optimal window step length; obtaining window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points; obtaining window fitness according to the data change characteristics in the adjacent optimal window step length range; and obtaining an adaptive window adjustment value according to the window adaptation degree and the window adjustment step length. According to the invention, an adaptive window area and a denoising temperature data sequence are obtained according to an adaptive window adjustment value and an optimal window step length, and environmental temperature early warning is carried out according to the denoising temperature data sequence; the denoising accuracy and the early warning accuracy are improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an intelligent early warning system for the ambient temperature of a large-caliber Internet of things water meter.
Background
The rapid development of the Internet of things technology introduces more intellectualization and automation in the water meter industry, and the large-caliber water meter is widely applied to industry and commerce; however, in extreme environments such as extremely cold or high temperature, monitoring of conventional water meters faces a series of challenges, and in order to ensure stable operation of the large-caliber internet of things water meter, temperature data prediction needs to be performed according to temperature data change characteristics.
The acquired real-time environmental temperature data of the large-caliber Internet of things water meter may contain different noise data, so that the accuracy of data prediction is low, and denoising processing is required to be performed on the acquired data before the data prediction. Conventional data denoising generally adopts a moving average method, and the method reduces fluctuation of data by calculating an average value of the data in a certain window; the accuracy of the denoising effect of the method depends on the value of the size of a window, a larger window can smooth the trend, but the change adaptability of data is slower, and the denoising effect is poor due to a smaller window. Therefore, the existing moving average method is difficult to accurately denoise and reflect the data change trend, so that the early warning accuracy of the ambient temperature is low.
Disclosure of Invention
In order to solve the technical problems that the prior moving average method is difficult to accurately remove noise and reflect data change trend, and the early warning accuracy of the environmental temperature is low, the invention aims to provide an environmental temperature intelligent early warning system of a large-caliber Internet of things water meter, and the adopted technical scheme is as follows:
the data acquisition module is used for acquiring a temperature data sequence for monitoring the ambient temperature of the water meter;
The window processing module is used for obtaining an initial window step length and an initial sampling sequence according to the data change characteristics of the temperature data sequence; selecting the initial window step length according to the data difference characteristic in the initial sampling sequence to obtain an optimal window step length;
the window analysis module is used for obtaining window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points; obtaining window fitness of the data point according to the data change characteristics in the adjacent optimal window step range of the data point; obtaining an adaptive window adjustment value of a data point according to the window fitness and the window adjustment step length;
The temperature early warning module is used for obtaining an adaptive window area of a data point according to the adaptive window adjustment value and the optimal window step length; obtaining a denoising temperature data sequence according to the data characteristics in the self-adaptive window area of the data point; and carrying out environmental temperature early warning according to the denoising temperature data sequence.
Further, the step of obtaining an initial window step size and an initial sampling sequence according to the data change characteristic of the temperature data sequence comprises the following steps:
sampling the temperature data sequence at length intervals of a preset step length to obtain a plurality of temperature sampling sequences of the preset step length; calculating the number of the numerical values with the least occurrence frequency in the temperature sampling sequence to obtain the minimum frequency characterization value; calculating and normalizing the difference value between the quantity corresponding to the mode in the temperature sampling sequence and the minimum frequency characterization value to obtain a period difference characteristic value of the temperature sampling sequence; calculating the average value of the cycle difference characteristic values of a plurality of temperature sampling sequences with preset step length to obtain the suitability of the preset step length;
selecting a preset step length corresponding to the maximum value of the suitability as the initial window step length; and taking the temperature sampling sequence corresponding to the initial window step length as the initial sampling sequence.
Further, the step of selecting the initial window step according to the data difference characteristic in the initial sampling sequence to obtain an optimal window step includes:
Calculating the numerical value difference of adjacent data points in the initial sampling sequence to obtain adjacent data difference; calculating variance of the adjacent data difference in the initial sampling sequence and carrying out negative correlation mapping to obtain a sequence stability characteristic value; calculating the product of the sequence stability characteristic value and the corresponding period difference characteristic value of the initial sampling sequence to obtain the window periodicity of the initial sampling sequence; calculating the average value of the window periodicity of the initial sampling sequence corresponding to the initial window step length to obtain the period fitness of the initial window step length;
And selecting an initial window step length corresponding to the maximum value of the period fitness as the optimal window step length.
Further, the step of obtaining a window adjustment step according to the data fluctuation characteristics within the optimal window step range of the data points comprises the following steps:
Taking the data point as a starting point in the temperature data sequence, taking the length range of the optimal window step length as a neighborhood region of the data point, calculating the variance of the data value in the neighborhood region of the data point, and normalizing to obtain window adjustment weight; when the window adjustment weight does not exceed a preset threshold, the window adjustment weight is a preset first constant; and calculating the product of the window adjustment weight and the optimal window step length, and rounding downwards to obtain the window adjustment step length of the data point.
Further, the step of obtaining window fitness of the data point according to the data change characteristics within the adjacent optimal window step range of the data point comprises the following steps:
Calculating the variance of a difference sequence of data in the neighborhood region of any data point in the temperature data sequence, and obtaining a local fluctuation characteristic value of the neighborhood region of any data point; calculating the average value of local fluctuation characteristic values of the neighborhood region of the data point and the neighborhood region adjacent to the front neighborhood region and the rear neighborhood region to obtain a local fluctuation average value; and calculating the difference value of the local fluctuation characteristic value corresponding to the data point and the local fluctuation average value, and carrying out negative correlation mapping to obtain the window fitness of the data point.
Further, the step of obtaining an adaptive window adjustment value for a data point based on the window fitness and the window adjustment step size comprises:
When the window fitness of the data point is lower than a preset fitness threshold, calculating a difference value between the preset fitness threshold and the window fitness to obtain a fitness difference degree; calculating the ratio of the adaptation difference degree to the preset adaptation threshold value to obtain an adjustment coefficient; calculating the difference value of the optimal window step length and the window adjustment step length of the data point to obtain a window adjustment reference; calculating the product of the adjustment coefficient and the window adjustment reference to obtain a window correction value of the data point;
When the window fitness of the data point is not lower than the preset fitness threshold, the window correction value of the data point is a preset second constant;
and calculating the sum of the window correction value and the window adjustment step length of the data point, and rounding down to obtain an adaptive window adjustment value of the data point.
Further, the step of obtaining an adaptive window area of data points from the adaptive window adjustment value and the optimal window step size comprises:
Calculating the sum of the optimal window step length and the adaptive window adjustment value of the data point to obtain the adaptive window length of the data point; and constructing an adaptive window area of the data points by taking the data points as a starting point and taking the data points in the adaptive window length in the temperature data sequence.
Further, the step of obtaining a de-noised temperature data sequence from the data characteristics in the adaptive window region of data points includes:
Calculating a numerical average value in the self-adaptive window area as a denoising characteristic value of the data point; and traversing to obtain the denoising characteristic value of each data point in the temperature data sequence, and constructing the denoising temperature data sequence.
The invention has the following beneficial effects:
In the invention, the change period characteristics of the temperature data sequence can be preliminarily determined by acquiring the initial window step length and the initial sampling sequence, and the period length of the temperature data sequence can be reflected by acquiring the optimal window step length, so that the initial window size in the denoising process is determined; the window adjustment step length can be obtained to adjust the size of the window according to the fluctuation characteristics of the data points, so that the denoising accuracy is primarily improved. The window fitness can be obtained to judge the fluctuation range of the data according to the data change characteristics of the area near the data point, and further judge the window size required by denoising. The self-adaptive window adjustment value can be obtained accurately according to the data fluctuation characteristics of the area near the data point, and the denoising accuracy is further improved. Finally, noise data can be accurately removed by obtaining the denoising temperature data sequence according to the self-adaptive window region, and the accuracy of the environmental temperature early warning is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of an intelligent early warning system for environmental temperature of a large-caliber internet of things water meter according to an embodiment of the invention.
Detailed Description
In order to further explain the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of the specific implementation, structure, characteristics and effects of the intelligent early warning system for the ambient temperature of the large-caliber internet of things water meter according to the invention, which is provided by the invention, with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of an intelligent environment temperature early warning system for a large-caliber Internet of things water meter, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of an intelligent early warning system for environmental temperature of a large-caliber internet of things water meter according to an embodiment of the invention is shown, the system comprises the following modules:
The data acquisition module S1 is configured to acquire a temperature data sequence for monitoring the ambient temperature of the water meter.
In the embodiment of the invention, the implementation scene is to denoise and early warn the data of the environmental temperature of the large-caliber Internet of things water meter; firstly, acquiring a temperature data sequence of the ambient temperature of the water meter, constructing the ambient temperature data of each acquisition time into the temperature data sequence according to a time sequence, and enabling an implementer to automatically determine the frequency of data acquisition according to an implementation scene.
The window processing module S2 is used for obtaining an initial window step length and an initial sampling sequence according to the data change characteristics of the temperature data sequence; and selecting the initial window step length according to the data difference characteristic in the initial sampling sequence to obtain the optimal window step length.
When the traditional moving average method is adopted for denoising, the denoising effect is determined by selecting a denoising window, the larger window can smooth the data change trend, early warning is untimely, and the denoising effect of the smaller window is poor. If the data accords with the periodical change trend, the periodical length of the data can be used as the window size of the moving average, so that the denoising effect is good; therefore, firstly, the periodic characteristics of the temperature data sequence are required to be obtained, and the initial window step length and the initial sampling sequence are obtained according to the data change characteristics of the temperature data sequence.
Preferably, in one embodiment of the present invention, the obtaining the initial window step size and the initial sampling sequence comprises: sampling the temperature data sequences at length intervals of preset step sizes to obtain a plurality of temperature sampling sequences of the preset step sizes; the implementer can determine the value range of the preset step length by himself according to the implementation scene, for example, when the preset step length is 3, sampling is performed in the temperature data sequence at intervals of 3 as length, starting from the first data, sampling the 4 th data point after 3 data intervals, traversing the temperature data sequence, and obtaining the first temperature sampling sequence; then starting from the second data, sampling the 5 th data point after 3 data intervals, traversing the temperature data sequence, and obtaining a second temperature sampling sequence; similarly, starting from the third data, a third temperature sample sequence is obtained, and a total of 3 temperature sample sequences are obtained. When the value of the preset step length is close to the period length of the temperature data sequence, the data point value in each temperature sampling sequence is close to the value of the data point in each temperature sampling sequence; calculating the number of the numerical values with the least occurrence frequency in the temperature sampling sequence to obtain the minimum frequency characterization value; calculating and normalizing the difference value between the quantity corresponding to the mode in the temperature sampling sequence and the minimum frequency characterization value to obtain a period difference characteristic value of the temperature sampling sequence; if the current preset step length is more consistent with the period length of the temperature data sequence, the more similar the data in the temperature sampling sequence are, the more the mode number is corresponding, the smaller the minimum frequency characterization value is, and the larger the period difference characteristic value is. Calculating the average value of the cycle difference characteristic values of a plurality of temperature sampling sequences with preset step length to obtain the suitability of the preset step length; when the cycle difference characteristic values of all the temperature sampling sequences under the preset step length are larger, the suitability is larger, which means that the length of the preset step length is more consistent with the cycle length of the temperature data sequence. Selecting a preset step length corresponding to the maximum value of the suitability as an initial window step length; and taking the temperature sampling sequence corresponding to the initial window step length as an initial sampling sequence.
Further, since a plurality of initial window steps with different sizes may exist at the same time, further analysis of the initial sampling sequence is required, and a change period of the most suitable temperature data sequence is selected, so that the denoising effect of the moving average is improved; therefore, selecting the initial window step length according to the data difference characteristic in the initial sampling sequence to obtain the optimal window step length; the method specifically comprises the following steps: calculating the numerical value difference of adjacent data points in the initial sampling sequence to obtain adjacent data difference; calculating variance of adjacent data difference in the initial sampling sequence and carrying out negative correlation mapping to obtain a sequence stability characteristic value; when the preset step length corresponding to the initial sampling sequence accords with the periodic characteristics of the temperature data sequence, the smaller and more similar the data difference in the initial sampling sequence is, the smaller the variance of the adjacent data difference is, and the larger the sequence stability characteristic value is. Calculating the product of the sequence stability characteristic value and the corresponding period difference characteristic value of the initial sampling sequence to obtain the window periodicity of the initial sampling sequence; when the sequence stability characteristic value and the period difference characteristic value are larger, the data in the initial sampling sequence are closer, the corresponding preset step length is more consistent with the period characteristic of the temperature data sequence, and the window periodicity is larger. Calculating the average value of the window periodicity of the initial sampling sequence corresponding to the initial window step length to obtain the period fitness of the initial window step length, wherein the larger the period fitness is, the more the initial window step length accords with the period length of the temperature data sequence; selecting an initial window step length corresponding to the maximum value of the period fitness as an optimal window step length; the optimal window step size best conforms to the cycle length characteristics of the temperature data sequence.
The window analysis module S3 is used for obtaining window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points; obtaining window fitness of the data point according to the data change characteristics in the adjacent optimal window step range of the data point; and obtaining an adaptive window adjustment value of the data point according to the window adaptation degree and the window adjustment step length.
The obtained optimal window step length is used as an initial window length in a moving average algorithm, but in order to improve the accuracy of data denoising, the window size needs to be adaptively adjusted, when the fluctuation of data near a certain data point is more obvious, the data near the data point is difficult to present periodicity, and larger window is needed for average denoising; the window adjustment step size is obtained based on the data fluctuation characteristics within the optimal window step size range of the data points.
Preferably, in one embodiment of the present invention, the acquiring the window adjustment step size includes: taking a data point as a starting point in a temperature data sequence, taking the length range of the optimal window step length as a neighborhood region of the data point, calculating the variance of the data value in the neighborhood region of the data point, and normalizing to obtain window adjustment weight; when the variance of the data value in the neighborhood region of the data point is larger, which means that the fluctuation is more obvious, a larger window is needed for denoising, and the window adjustment weight is larger. When the window adjustment weight does not exceed the preset threshold value, the window adjustment weight is a preset first constant; in the invention, the preset threshold value is 0.3, and the preset first constant is 0, which means that the data fluctuation condition in the neighborhood region of the data point is normal, and an oversized window is not needed at the moment; the practitioner can set the super parameters by himself according to the implementation scene. And calculating the product of the window adjustment weight and the optimal window step length, rounding downwards to obtain the window adjustment step length of the data point, and when the fluctuation characteristic is more obvious, the window adjustment weight is larger, the window to be added is larger, and the window adjustment step length is longer.
Further, if the fluctuation characteristic of the neighborhood region of the data point is more obvious than that of the surrounding region, the data in the neighborhood region of the data point is difficult to accurately denoise only, and in order to improve the denoising accuracy, not only other data points in the window adjustment step length range, but also larger windows are needed to denoise the data points; if the fluctuation characteristics of the neighborhood region of the data point are similar to those of the adjacent neighborhood region, which means that the fluctuation characteristics of the whole temperature data sequence are obvious, a larger window is not needed; the window fitness of the data point is obtained based on the data change characteristics within the range of adjacent optimal window steps of the data point.
Preferably, in one embodiment of the present invention, the obtaining window fitness includes: calculating the variance of a differential sequence of data in a neighborhood region of any data point in the temperature data sequence, and obtaining a local fluctuation characteristic value of the neighborhood region of the any data point; it should be noted that, the acquisition of the differential sequence belongs to the prior art, and specific acquisition steps are not repeated; the more complex the data fluctuations in the neighborhood region, the larger the local fluctuation feature value. Calculating the average value of local fluctuation characteristic values of a neighborhood region of a data point and a neighborhood region adjacent to the neighborhood region before and after to obtain a local fluctuation average value; the local fluctuation average reflects the data fluctuation characteristics of a larger area around the data point, and the larger the local fluctuation average is, the more obviously complex the data fluctuation in the larger area around the data point is. Calculating the difference value between the local fluctuation characteristic value corresponding to the data point and the local fluctuation average value and carrying out negative correlation mapping to obtain the window fitness of the data point; when the local characteristic fluctuation value corresponding to the data point is smaller, the window fitness is larger, and the window for denoising the data point does not need to be continuously increased; conversely, when the window fitness is smaller, which means that the local fluctuation characteristic value of the data point is larger than the local fluctuation average value, in order to improve the denoising accuracy of the data point, a larger denoising window is required.
Further, the adaptive window adjustment value of the data point can be obtained according to the window fitness and the window adjustment step length, which specifically includes: when the window fitness of the data point is lower than a preset fitness threshold, calculating a difference value between the preset fitness threshold and the window fitness to obtain a fitness difference degree; when the window fitness is smaller, the adaptability difference is larger, and in the embodiment of the invention, the preset adaptability threshold is 0.5, and an implementer can determine according to implementation scenes. Calculating the ratio of the adaptation difference degree to a preset adaptation threshold value to obtain an adjustment coefficient; when the local fluctuation characteristic value of the data point is larger, the local fluctuation characteristic is more obvious, the adaptation difference degree is larger, the adjustment coefficient is larger, and a larger window is needed for denoising. Calculating the difference value of the optimal window step length and the window adjustment step length of the data point to obtain a window adjustment reference; the maximum value of the window adjustment reference does not exceed the optimal window step length, the window size after final adjustment is ensured to be 2 times of the optimal window step length, namely, two change periods in the temperature data sequence are not exceeded, and the denoising error is reduced. Calculating the product of the adjustment coefficient and the window adjustment reference to obtain a window correction value of the data point; the larger the adjustment coefficient of the data point, the larger the window correction value. When the window fitness of the data point is not lower than a preset adaptive threshold, meaning that larger window denoising is not needed, the window correction value of the data point is a preset second constant; in the embodiment of the invention, the second constant is preset to be 0, and an implementer can determine according to implementation scenes. And calculating the sum of the window correction value and the window adjustment step length of the data point, and rounding down to obtain an adaptive window adjustment value of the data point, wherein when the local characteristic fluctuation of the data point is more obvious, the adaptive window adjustment value is larger. When the fitness of the data point is lower than a preset fitness threshold, the formula for obtaining the adaptive window adjustment value comprises:
in the method, in the process of the invention, Adaptive window adjustment value representing data point,/>Window adjustment step size representing the data point,/>Representing the optimal window step size,/>Representing a preset adaptation threshold,/>Window fitness representing the data point,/>Representing fitness differences,/>Representing adjustment coefficient,/>Representing window adjustment references,/>Window correction representing data point,/>Representing a downward rounding function.
The temperature early warning module S4 is used for obtaining an adaptive window area of a data point according to the adaptive window adjustment value and the optimal window step length; obtaining a denoising temperature data sequence according to the data characteristics in the self-adaptive window area of the data point; and carrying out environmental temperature early warning according to the denoising temperature data sequence.
After the adaptive window adjustment value of the data point is obtained, the adaptive window area of the data point can be obtained according to the adaptive window adjustment value and the optimal window step length, which specifically comprises the following steps: calculating the sum of the optimal window step length and the self-adaptive window adjustment value of the data point to obtain the self-adaptive window length of the data point; an adaptive window region for a data point within the adaptive window length is constructed from the data point starting from the data point in the temperature data sequence.
Further, a denoising temperature data sequence can be obtained according to the data characteristics in the self-adaptive window area of the data point, and the numerical average value in the self-adaptive window area is calculated and used as the denoising characteristic value of the data point; and traversing to obtain the denoising characteristic value of each data point in the temperature data sequence, and constructing a denoising temperature data sequence. Finally, the environmental temperature early warning can be carried out according to the denoising temperature data sequence, in the embodiment of the invention, the denoising temperature data sequence is predicted through the neural network, the future temperature change trend is obtained, the temperature threshold value is set for early warning, and the fact that the neural network predicted data sequence belongs to the prior art is not repeated in specific calculation steps is needed; the enforcer can determine the early warning method according to the denoising temperature data sequence according to the implementation scene by himself, and the method is not limited herein. So far, denoising is carried out according to the self-adaptive window area of the data points, so that the denoising accuracy and the early warning accuracy of each data point can be improved.
In summary, the embodiment of the invention provides an ambient temperature intelligent early warning system of a large-caliber internet of things water meter; obtaining an initial window step length and an initial sampling sequence according to the data change characteristics of the temperature data sequence; selecting an initial window step length according to the initial sampling sequence to obtain an optimal window step length; obtaining window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points; obtaining window fitness according to the data change characteristics in the adjacent optimal window step length range; and obtaining an adaptive window adjustment value according to the window adaptation degree and the window adjustment step length. According to the invention, an adaptive window area and a denoising temperature data sequence are obtained according to an adaptive window adjustment value and an optimal window step length, and environmental temperature early warning is carried out according to the denoising temperature data sequence; the denoising accuracy and the early warning accuracy are improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (5)
1. An ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge, its characterized in that, the system includes following module:
the data acquisition module is used for acquiring a temperature data sequence for monitoring the ambient temperature of the water meter;
The window processing module is used for obtaining an initial window step length and an initial sampling sequence according to the data change characteristics of the temperature data sequence; selecting the initial window step length according to the data difference characteristic in the initial sampling sequence to obtain an optimal window step length;
the window analysis module is used for obtaining window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points; obtaining window fitness of the data point according to the data change characteristics in the adjacent optimal window step range of the data point; obtaining an adaptive window adjustment value of a data point according to the window fitness and the window adjustment step length;
The temperature early warning module is used for obtaining an adaptive window area of a data point according to the adaptive window adjustment value and the optimal window step length; obtaining a denoising temperature data sequence according to the data characteristics in the self-adaptive window area of the data point; performing environmental temperature early warning according to the denoising temperature data sequence;
The step of obtaining the window adjustment step length according to the data fluctuation characteristics in the optimal window step length range of the data points comprises the following steps:
taking the data point as a starting point in the temperature data sequence, taking the length range of the optimal window step length as a neighborhood region of the data point, calculating the variance of the data value in the neighborhood region of the data point, and normalizing to obtain window adjustment weight; when the window adjustment weight does not exceed a preset threshold, the window adjustment weight is a preset first constant; calculating the product of the window adjustment weight and the optimal window step length, and rounding downwards to obtain the window adjustment step length of the data point;
The step of obtaining window fitness of the data point according to the data change characteristics in the adjacent optimal window step range of the data point comprises the following steps:
Calculating the variance of a difference sequence of data in the neighborhood region of any data point in the temperature data sequence, and obtaining a local fluctuation characteristic value of the neighborhood region of any data point; calculating the average value of local fluctuation characteristic values of the neighborhood region of the data point and the neighborhood region adjacent to the front neighborhood region and the rear neighborhood region to obtain a local fluctuation average value; calculating the difference value of the local fluctuation characteristic value corresponding to the data point and the local fluctuation average value and carrying out negative correlation mapping to obtain window fitness of the data point;
the step of obtaining an adaptive window adjustment value for a data point based on the window fitness and the window adjustment step size comprises:
When the window fitness of the data point is lower than a preset fitness threshold, calculating a difference value between the preset fitness threshold and the window fitness to obtain a fitness difference degree; calculating the ratio of the adaptation difference degree to the preset adaptation threshold value to obtain an adjustment coefficient; calculating the difference value of the optimal window step length and the window adjustment step length of the data point to obtain a window adjustment reference; calculating the product of the adjustment coefficient and the window adjustment reference to obtain a window correction value of the data point;
When the window fitness of the data point is not lower than the preset fitness threshold, the window correction value of the data point is a preset second constant;
and calculating the sum of the window correction value and the window adjustment step length of the data point, and rounding down to obtain an adaptive window adjustment value of the data point.
2. The intelligent early warning system for environmental temperature of a large-caliber internet of things water meter according to claim 1, wherein the step of obtaining an initial window step size and an initial sampling sequence according to the data change characteristic of the temperature data sequence comprises the following steps:
sampling the temperature data sequence at length intervals of a preset step length to obtain a plurality of temperature sampling sequences of the preset step length; calculating the number of the numerical values with the least occurrence frequency in the temperature sampling sequence to obtain the minimum frequency characterization value; calculating and normalizing the difference value between the quantity corresponding to the mode in the temperature sampling sequence and the minimum frequency characterization value to obtain a period difference characteristic value of the temperature sampling sequence; calculating the average value of the cycle difference characteristic values of a plurality of temperature sampling sequences with preset step length to obtain the suitability of the preset step length;
selecting a preset step length corresponding to the maximum value of the suitability as the initial window step length; and taking the temperature sampling sequence corresponding to the initial window step length as the initial sampling sequence.
3. The intelligent early warning system for environmental temperature of a large-caliber internet of things water meter according to claim 2, wherein the step of selecting the initial window step length according to the data difference characteristic in the initial sampling sequence to obtain an optimal window step length comprises the following steps:
Calculating the numerical value difference of adjacent data points in the initial sampling sequence to obtain adjacent data difference; calculating variance of the adjacent data difference in the initial sampling sequence and carrying out negative correlation mapping to obtain a sequence stability characteristic value; calculating the product of the sequence stability characteristic value and the corresponding period difference characteristic value of the initial sampling sequence to obtain the window periodicity of the initial sampling sequence; calculating the average value of the window periodicity of the initial sampling sequence corresponding to the initial window step length to obtain the period fitness of the initial window step length;
And selecting an initial window step length corresponding to the maximum value of the period fitness as the optimal window step length.
4. The intelligent early warning system for environmental temperature of a large-caliber internet of things water meter according to claim 1, wherein the step of obtaining the adaptive window area of the data point according to the adaptive window adjustment value and the optimal window step length comprises:
Calculating the sum of the optimal window step length and the adaptive window adjustment value of the data point to obtain the adaptive window length of the data point; and constructing an adaptive window area of the data points by taking the data points as a starting point and taking the data points in the adaptive window length in the temperature data sequence.
5. The intelligent early warning system for ambient temperature of a large-caliber internet of things water meter according to claim 4, wherein the step of obtaining the de-noised temperature data sequence according to the data characteristics in the adaptive window area of the data points comprises the steps of:
Calculating a numerical average value in the self-adaptive window area as a denoising characteristic value of the data point; and traversing to obtain the denoising characteristic value of each data point in the temperature data sequence, and constructing the denoising temperature data sequence.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410210577.9A CN117786325B (en) | 2024-02-27 | 2024-02-27 | Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410210577.9A CN117786325B (en) | 2024-02-27 | 2024-02-27 | Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117786325A CN117786325A (en) | 2024-03-29 |
CN117786325B true CN117786325B (en) | 2024-04-30 |
Family
ID=90389493
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410210577.9A Active CN117786325B (en) | 2024-02-27 | 2024-02-27 | Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117786325B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117970859B (en) * | 2024-04-02 | 2024-06-21 | 山西顺达胜业通信工程有限公司 | Intelligent control method for equipment under low-power-consumption wide area Internet of things |
CN118197635B (en) * | 2024-05-16 | 2024-07-09 | 尽开科技(大连)有限公司 | Beef cattle health detection data processing method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021258832A1 (en) * | 2020-06-23 | 2021-12-30 | 青岛科技大学 | Method for denoising underwater acoustic signal on the basis of adaptive window filtering and wavelet threshold optimization |
CN115840893A (en) * | 2022-12-09 | 2023-03-24 | 北京数洋智慧科技有限公司 | Multivariable time series prediction method and device |
WO2023105277A1 (en) * | 2021-12-09 | 2023-06-15 | Sensetime International Pte. Ltd. | Data sampling method and apparatus, and storage medium |
CN116578041A (en) * | 2023-06-05 | 2023-08-11 | 浙江德欧电气技术股份有限公司 | Data processing method for CNC controller |
CN116702081A (en) * | 2023-08-07 | 2023-09-05 | 西安格蒂电力有限公司 | Intelligent inspection method for power distribution equipment based on artificial intelligence |
CN117407700A (en) * | 2023-12-14 | 2024-01-16 | 国网山东省电力公司莱芜供电公司 | Method for monitoring working environment in live working process |
CN117496359A (en) * | 2023-12-29 | 2024-02-02 | 浙江大学山东(临沂)现代农业研究院 | Plant planting layout monitoring method and system based on three-dimensional point cloud |
CN117559448A (en) * | 2024-01-12 | 2024-02-13 | 山东德源电力科技股份有限公司 | Power consumption load prediction analysis method and system for special transformer acquisition terminal |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7818224B2 (en) * | 2001-03-22 | 2010-10-19 | Boerner Sean T | Method and system to identify discrete trends in time series |
TW202403912A (en) * | 2022-07-01 | 2024-01-16 | 聯華電子股份有限公司 | Fault detection method for detecting behavior deviation of parameters |
-
2024
- 2024-02-27 CN CN202410210577.9A patent/CN117786325B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021258832A1 (en) * | 2020-06-23 | 2021-12-30 | 青岛科技大学 | Method for denoising underwater acoustic signal on the basis of adaptive window filtering and wavelet threshold optimization |
WO2023105277A1 (en) * | 2021-12-09 | 2023-06-15 | Sensetime International Pte. Ltd. | Data sampling method and apparatus, and storage medium |
CN115840893A (en) * | 2022-12-09 | 2023-03-24 | 北京数洋智慧科技有限公司 | Multivariable time series prediction method and device |
CN116578041A (en) * | 2023-06-05 | 2023-08-11 | 浙江德欧电气技术股份有限公司 | Data processing method for CNC controller |
CN116702081A (en) * | 2023-08-07 | 2023-09-05 | 西安格蒂电力有限公司 | Intelligent inspection method for power distribution equipment based on artificial intelligence |
CN117407700A (en) * | 2023-12-14 | 2024-01-16 | 国网山东省电力公司莱芜供电公司 | Method for monitoring working environment in live working process |
CN117496359A (en) * | 2023-12-29 | 2024-02-02 | 浙江大学山东(临沂)现代农业研究院 | Plant planting layout monitoring method and system based on three-dimensional point cloud |
CN117559448A (en) * | 2024-01-12 | 2024-02-13 | 山东德源电力科技股份有限公司 | Power consumption load prediction analysis method and system for special transformer acquisition terminal |
Non-Patent Citations (4)
Title |
---|
TCP_SDF: Transport Control to Achieve Smooth Data Flow at a Stable Throughput;Xukang Lu等;2013 IEEE 10th International Conference on High Performance Computing and Communications & 2013 IEEE International Conference on Embedded and Ubiquitous Computing;20140612;全文 * |
基于双约束滑动时间窗口的告警预处理方法研究;李彤岩;李兴明;;计算机应用研究;20130215(02);全文 * |
基于每时晴空指数的大规模光伏电站出力多维时间序列模拟;李国庆;李欣彤;边竞;牛继涛;丁煌;陈卫东;;电网技术;20200930(09);全文 * |
自适应多尺度窗口平均光谱平滑;季江;高鹏飞;贾南南;杨蕊;郭汉明;瑚琦;庄松林;;光谱学与光谱分析;20150515(05);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN117786325A (en) | 2024-03-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117786325B (en) | Ambient temperature wisdom early warning system of heavy-calibre thing networking water gauge | |
CN116702081B (en) | Intelligent inspection method for power distribution equipment based on artificial intelligence | |
CN116828070B (en) | Intelligent power grid data optimization transmission method | |
CN117407700B (en) | Method for monitoring working environment in live working process | |
CN110852515A (en) | Water quality index prediction method based on mixed long-time and short-time memory neural network | |
CN117574061B (en) | PM2.5 and ozone pollution cooperative prevention and control prediction method and system | |
CN113177633A (en) | Deep decoupling time sequence prediction method | |
CN117454085B (en) | Vehicle online monitoring method and system | |
CN116418882B (en) | Memory data compression method based on HPLC dual-mode carrier communication | |
CN117196353A (en) | Environmental pollution assessment and monitoring method and system based on big data | |
CN112559598B (en) | Telemetry time series data abnormity detection method and system based on graph neural network | |
CN114358435A (en) | Pollution source-water quality prediction model weight influence calculation method of two-stage space-time attention mechanism | |
CN117607731B (en) | Full-color LED electronic circuit display screen power failure detection method | |
CN111931354A (en) | Transformer top layer oil temperature prediction method based on gray-autoregressive differential moving average model | |
CN117236084A (en) | Intelligent management method and system for woodworking machining production | |
CN114186596B (en) | Multi-window identification method and device for spectrogram peaks and electronic equipment | |
CN115314412A (en) | Operation and maintenance-oriented type-adaptive index prediction early warning method and device | |
CN116582596A (en) | Data processing method and device and PLC data transmission system based on Internet of things | |
CN117541582A (en) | IGBT insulation quality detection method for high-frequency converter | |
CN115099291A (en) | Building energy-saving monitoring method | |
CN118395091A (en) | Data efficient acquisition system based on everything interconnection | |
CN118134539B (en) | User behavior prediction method based on intelligent kitchen multi-source data fusion | |
CN117874685B (en) | Pressure data abnormality early warning method for direct-current variable-frequency electric half ball valve | |
CN117692012B (en) | Remote monitoring and transmitting method for temperature data of intelligent sleeping bag | |
CN118090078B (en) | Leakage online monitoring method for closed circulation water cooling system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |