CN116680661B - Multi-dimensional data-based automatic gas regulator pressure monitoring method - Google Patents
Multi-dimensional data-based automatic gas regulator pressure monitoring method Download PDFInfo
- Publication number
- CN116680661B CN116680661B CN202310968144.5A CN202310968144A CN116680661B CN 116680661 B CN116680661 B CN 116680661B CN 202310968144 A CN202310968144 A CN 202310968144A CN 116680661 B CN116680661 B CN 116680661B
- Authority
- CN
- China
- Prior art keywords
- data
- difference
- value
- parameter data
- predicted
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012544 monitoring process Methods 0.000 title claims abstract description 36
- 230000001105 regulatory effect Effects 0.000 claims abstract description 76
- 238000013507 mapping Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 10
- 230000000875 corresponding effect Effects 0.000 description 39
- 239000007789 gas Substances 0.000 description 37
- 230000000694 effects Effects 0.000 description 14
- 239000002737 fuel gas Substances 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L11/00—Measuring steady or quasi-steady pressure of a fluid or a fluent solid material by means not provided for in group G01L7/00 or G01L9/00
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Feedback Control In General (AREA)
- Control Of Fluid Pressure (AREA)
Abstract
The invention relates to the technical field of data prediction, in particular to a gas automatic regulator cubicle pressure monitoring method based on multidimensional data. According to the method, an integrated moving average autoregressive model is built through historical data, and autoregressive coefficients are built in the integrated moving average autoregressive model through data value correlation and change trend correlation; and obtaining an error coefficient according to the distribution difference and the numerical value difference between the predicted data and the actual data. And obtaining adjustment coefficients according to variation trend differences and data differences between the predicted values and the actual values at each moment in the adjacent time range corresponding to the obtained predicted pressure regulating parameter data, optimizing, and monitoring the pressure of the automatic gas regulator according to the optimized pressure regulating parameter data in multiple dimensions. According to the invention, a high-efficiency integrated moving average autoregressive model is constructed, prediction data is optimized, optimized pressure regulating parameter data with excellent accuracy is obtained, and timely monitoring of the pressure of the automatic gas regulator is realized.
Description
Technical Field
The invention relates to the technical field of data prediction, in particular to a gas automatic regulator cubicle pressure monitoring method based on multidimensional data.
Background
The automatic gas regulator cabinet can automatically regulate and control the pressure of gas so as to adapt to the requirements of different users and implementation scenes. The gas is carried by high-pressure seal pipeline, and the gas pressure in the pipeline needs to be guaranteed to the gas automatic pressure regulating cabinet in normal within range, consequently in order to guarantee the safety of gas pipeline, can monitor the gas parameter in the pipeline through sensor equipment, judges whether corresponding parameter is in unsafe within range according to the threshold value that sets up, and then utilizes the gas automatic pressure regulating cabinet to carry out pressure regulation. However, in actual situations, if an abnormal gas pressure is detected at a certain moment, the pipeline is damaged at the moment, so that the safety of the pipeline cannot be ensured, and in order to avoid the problem, the pressure monitoring needs to be performed at each moment in advance.
In the prior art, in order to realize the advanced accurate monitoring of the pressure of the automatic gas regulator, a data prediction method is adopted to predict the data at the future moment, so that the advanced monitoring of the pressure of the gas is realized. However, in the prediction process in the prior art, the fact that the gas data present different data characteristics along with different time periods is not considered, so that the prediction data is inaccurate, namely, the pressure monitoring of the gas automatic pressure regulating cabinet is inaccurate, and the pressure regulating effect is affected.
Disclosure of Invention
In order to solve the technical problems that in the prior art, pressure data of fuel gas cannot be accurately predicted, and pressure monitoring is inaccurate, the invention aims to provide a method for monitoring the pressure of a fuel gas automatic pressure regulating cabinet based on multidimensional data, and the adopted technical scheme is as follows:
a method for monitoring the pressure of a gas automatic regulator cubicle based on multidimensional data, comprising the following steps:
acquiring pressure regulating parameter data of the automatic gas pressure regulating cabinet at each moment in each dimension in a preset time range; the preamble data on the time sequence of the voltage regulation parameter data at each moment is the historical parameter data at the corresponding moment;
constructing an integrated moving average autoregressive model according to the pressure regulating parameter data; the method for acquiring the autoregressive coefficients in the integrated moving average autoregressive model comprises the following steps: acquiring the data value correlation and the change trend correlation between the voltage regulation parameter data and the corresponding historical parameter data at each moment; obtaining an autoregressive coefficient at each moment according to the data value correlation and the change trend correlation; the error coefficient acquisition method in the integrated moving average autoregressive model comprises the following steps: obtaining the error coefficient according to the distribution difference and the numerical value difference between the predicted data and the actual data;
predicting predicted pressure regulating parameter data at a future moment by utilizing the integrated moving average autoregressive model according to the real-time pressure regulating parameter data; obtaining an adjustment coefficient according to the variation trend difference and the data difference between the predicted value and the actual data at each time in the preset adjacent time range before the corresponding time of the predicted voltage regulation parameter data; adjusting the predicted voltage regulation parameter data according to the adjustment coefficient to obtain optimized voltage regulation parameter data;
and the optimized pressure regulating parameter data in all dimensions form multidimensional optimized pressure regulating parameter data, and the pressure monitoring is carried out on the automatic gas pressure regulating cabinet according to the multidimensional optimized pressure regulating parameter data.
Further, the method for acquiring the data value correlation comprises the following steps:
acquiring a historical average data value of the historical parameter data corresponding to the voltage regulation parameter data at each moment, acquiring a first data value difference between the voltage regulation parameter data and the corresponding historical average data value, and acquiring the data value correlation according to the first data value difference, wherein the first data value difference and the data value correlation are in a negative correlation.
Further, the method for acquiring the correlation of the variation trend comprises the following steps:
mapping the voltage regulating parameter data in the preset time range to a two-dimensional coordinate system, wherein the abscissa of the two-dimensional coordinate system is moment, and the ordinate is a data value; obtaining the absolute value of the tangential slope of a data point at each moment in the two-dimensional coordinate system;
acquiring a first average tangential slope absolute value of all data points of each voltage regulation parameter data within a preset judging time range; acquiring second average tangential slope absolute values corresponding to all data points in the preset time range;
obtaining the change trend correlation according to a first slope absolute value difference between the first average tangential slope absolute value and the second average tangential slope absolute value; the first slope absolute value difference and the change trend correlation are in a negative correlation relationship.
Further, the data value correlation and the change trend correlation are in positive correlation with the autoregressive coefficient.
Further, the method for obtaining the error coefficient comprises the following steps:
setting an initial error coefficient to 0, and predicting data according to an integrated moving average autoregressive model corresponding to the initial error coefficient to obtain predicted data and a predicted data curve corresponding to the predicted data in the preset time range; obtaining actual data in the preset time range and an actual data curve corresponding to the actual data;
obtaining a data value fluctuation characteristic difference between the predicted data curve and the actual data curve; obtaining a first average data value difference between the predicted data and the actual data; and obtaining the error coefficient according to the data value fluctuation characteristic difference and the first average data value difference, wherein the data value fluctuation characteristic difference and the first average data value difference are in positive correlation with the error coefficient.
Further, the method for obtaining the adjustment coefficient includes:
acquiring second average data value differences between the predicted value and the actual data at all moments in the adjacent time range;
acquiring a third average tangential slope absolute value of all predicted values in the two-dimensional coordinate system at all moments in the adjacent time range; acquiring the absolute value of the fourth average tangential slope of all actual data in the two-dimensional coordinate system at all moments in the adjacent time range; acquiring a second slope absolute value difference between the third average tangential slope absolute value and the fourth average tangential slope absolute value;
and multiplying the second average data value difference by the second slope absolute value difference, and then carrying out normalization and negative correlation mapping to obtain the adjustment coefficient.
Further, the method for acquiring the optimized voltage regulation parameter data comprises the following steps:
multiplying the adjustment coefficients of the predicted voltage regulation parameter data field to obtain the optimized voltage regulation parameter data.
Further, the method for monitoring the pressure of the automatic gas regulator according to the multidimensional optimized pressure regulating parameter data further comprises the following steps:
calculating and normalizing the difference between the optimized voltage regulating parameter data in each dimension and a preset limit value in the corresponding dimension in the multidimensional optimized voltage regulating parameter data to obtain a state difference in each dimension; and carrying out pressure adjustment on the automatic gas regulator according to the relation between the average state difference under all dimensions and the preset judgment threshold value at real time.
Further, the judgment threshold is set to be 0.85, and when the state difference is larger than the judgment threshold, the automatic pressure regulating cabinet is subjected to pressure regulation at real time.
Further, the method for acquiring the state difference further comprises the following steps: and taking the ratio between the optimized voltage regulating parameter data in each dimension and the preset limit value in the corresponding dimension in the multidimensional optimized voltage regulating parameter data as the state difference.
The invention has the following beneficial effects:
according to the method, an integrated moving average autoregressive model is built based on data in a preset time range, and in the integrated moving average autoregressive model, the autoregressive coefficient of the integrated moving average autoregressive model is obtained through correlation of historical parameter data corresponding to each moment and data values and correlation of variation trends between the corresponding moments. Because the autoregressive coefficients take into account both the data values themselves and the data changes, the data predicted from the autocorrelation coefficients is more accurate. And further combining the error coefficient to obtain an integrated moving average autoregressive model with good prediction effect. According to the embodiment of the invention, the predicted voltage regulating parameters output according to the integrated moving average autoregressive model are further optimized, the adjusting coefficient for optimization is obtained according to the predicted difference in the adjacent time range of the corresponding moment of the predicted voltage regulating parameters, meanwhile, the data difference and the variation trend difference are considered, the quality of the current predicted effect is further determined, the optimized voltage regulating parameter data is obtained according to the adjusting coefficient, and the optimized voltage regulating parameter data with excellent predicted effect is obtained. And further, the accurate pressure monitoring of the automatic gas regulator is realized by combining multidimensional optimized pressure regulating parameter data.
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 flowchart of a method for monitoring the pressure of a gas automatic regulator cubicle based on multidimensional data according to an embodiment of the present 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 specific implementation, structure, characteristics and effects of the automatic gas regulator cubicle pressure monitoring method based on multidimensional data 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 a gas automatic pressure regulating cabinet pressure monitoring method based on multidimensional data, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring pressure of a gas automatic regulator cubicle based on multidimensional data according to an embodiment of the present invention is shown, where the method includes:
step S1: acquiring pressure regulating parameter data of the automatic gas pressure regulating cabinet at each moment in each dimension in a preset time range; the preamble data on the time sequence of the voltage regulation parameter data at each time is the historical parameter data at the corresponding time.
When the fuel gas is transmitted in the pipeline, various data types are reflected, namely, the fuel gas automatic pressure regulating cabinet corresponds to data in a plurality of dimensions, and in one embodiment of the invention, the data in the plurality of dimensions such as fuel gas pressure, fuel gas speed and fuel gas concentration are considered, so that timely and accurate predictive monitoring is required to be realized for each dimension, and further, the pressure regulation of the fuel gas in the pipeline by the fuel gas automatic pressure regulating cabinet is facilitated. In order to achieve accuracy of prediction, for the time to be predicted, statistics is carried out on pressure regulating parameter data in a preset time range before the time to be predicted as training data, and a prediction model is further constructed. Therefore, the embodiment of the invention needs to acquire the pressure regulating parameter data of the automatic gas pressure regulating cabinet at each time in each dimension within a preset time range, wherein the preamble data on the time sequence of the pressure regulating parameter data at each time is the historical parameter data at the corresponding time. It should be noted that, in other embodiments, other specific status parameters may be selected according to the specific status of the pipeline, which is not limited and described herein.
In the embodiment of the invention, the size is smallAnd the data in each dimension of the automatic gas regulator cabinet are collected as a unit of collection time, namely, every one hour. The preset time range is set to be one week, namely, the data of 168 hours before the current moment is counted to be used as data required by model construction. As an example, the initial time within the preset time range is set toReal time is set to +.>I.e. for real time instant +.>To->The data between moments is historical parameter data.
It should be noted that, in the embodiments of the present invention, the data prediction method in each dimension is the same, so for convenience of description, only parameter data in one dimension is described in the following description.
Step S2: and constructing an integrated moving average autoregressive model according to the pressure regulating parameter data.
And (3) constructing a prediction model based on the data obtained in the step S1. The integrated moving average autoregressive model has excellent prediction effect on time series data prediction, is a classical model for time series analysis and prediction, and each historical data corresponds to one autoregressive coefficient in the model, and the autoregressive coefficient can represent the linear relation between the corresponding historical data and the current time. Further determining random errors of each data point, obtaining error coefficients, and combining the error coefficients to realize accurate data prediction. In one embodiment of the invention the mathematical expression for integrating the moving average autoregressive model is:
wherein,,is->Predicted data value at time, +.>Is->Data value at time, +.>Is the firstData value at time, +.>Is->Autoregressive coefficients at time, < >>Is->Autoregressive coefficients at time, < >>For the initial time +.>Lower data value->For the initial time +.>Auto regression coefficient of->Is the error coefficient.
In order to realize accurate monitoring of the pressure data of the automatic gas regulator cubicle, accurate autoregressive coefficient determination is required to be carried out on the data at each moment in the integration moving average autoregressive model, and the prediction effect of the integration moving average autoregressive model can be effectively improved by determining the accurate autoregressive coefficient. Therefore, the data value correlation and the change trend correlation between the voltage-regulating parameter data and the corresponding historical parameter data at each moment need to be obtained, and the autoregressive coefficient at each moment is obtained according to the data value correlation and the change trend correlation, namely, the autoregressive coefficient simultaneously considers the difference and the change of the data at each moment compared with the prior moment, and the autoregressive coefficient of the final characterization linear relation is determined through the correlation in two dimensions.
Preferably, in one embodiment of the present invention, the method for acquiring the correlation of the data value includes:
the method comprises the steps of obtaining historical average data values of historical parameter data corresponding to voltage regulation parameter data at each moment, obtaining first data value differences between the voltage regulation parameter data and the corresponding historical average data values, and obtaining data value correlation according to the first data value differences, wherein the first data value differences and the data value correlation are in negative correlation, namely the larger the first data value differences are, the larger the differences between the voltage regulation parameter data at the current moment and the integral historical parameter data are, the weaker the data referential at the current moment is, and the weaker the correlation between the voltage regulation parameter data and the data at the moment to be predicted is. The formula for the correlation of data values in one embodiment of the invention is:
wherein,,is->Correlation of data values corresponding to the respective voltage regulation parameter data,/->Is->Data values corresponding to the individual voltage regulation parameter data, < >>Is->Data value of the individual history parameter data, +.>Is->The number of historical parameter data corresponding to the individual voltage regulation parameter data.
In the data value correlation formula,is the first data value difference, wherein +.>The effect of (2) is to prevent the denominator from being 0 and to implement the negative correlation mapping by taking the first data value difference as the denominator. In other embodiments of the present invention, the negative correlation mapping may be implemented by other basic mathematical methods, which are not described and limited herein.
Preferably, in one embodiment of the present invention, the method for acquiring the correlation of the trend of the change includes:
mapping the voltage regulating parameter data within a preset time range to a two-dimensional coordinate system, wherein the abscissa of the two-dimensional coordinate system is moment, and the ordinate is a data value; obtaining the absolute value of the tangential slope of a data point at each moment in a two-dimensional coordinate system, wherein the absolute value of the tangential slope represents the change of the data value at the current moment; in order to prevent inaccuracy caused by single data feature analysis at the current moment, further acquiring first average tangential slope absolute values of all data points of each voltage regulation parameter data within a preset judging time range by a local analysis method, wherein the first average tangential slope absolute values are used as data value change trend features at the current moment; acquiring absolute values of second average tangential slopes corresponding to all data points in a preset time range; the change trend correlation is obtained according to the first slope absolute value difference between the first average tangent slope absolute value and the second average tangent slope absolute value, the first slope absolute value difference and the change trend correlation form a negative correlation, namely, the larger the first slope absolute value difference is, the larger difference between the data change trend at the current moment and the change trend of the whole historical data is indicated, the poorer the reference property of the voltage regulating parameter data corresponding to the current moment is, and the weaker the corresponding change trend correlation is. In one embodiment of the present invention, the formula of the trend correlation is:
wherein,,is->Trend correlation of the pressure regulating parameter data at each moment, +.>As an exponential function based on natural constants, < +.>Is->Number of moments in the judgment time range of the individual moments, < >>To judge the +.>Tangential slope at each moment, +.>For a preset timeThe number of times in the range, +.>Is the +.>Tangential slope at each time instant. In one embodiment of the present invention, the judgment time range is set to 24 times, that is, 24 times before the current time constitute the judgment time range.
In the trend correlation formula, negative correlation mapping is realized through an exponential function based on a natural constant, and normalization is carried out, namely the obtained value range of the trend correlation is between 0 and 1.Is the absolute value of the first average tangential slope used for characterizing the +.>Trend of change of pressure regulating parameter data at each moment, < +.>The absolute value of the slope of the second average tangent is used for representing the overall data change trend in the whole time range, so that the absolute value difference of the first slope is obtained through the difference of the absolute value of the second average tangent and the overall data change trend, and further the change trend correlation is obtained.
Preferably, in one embodiment of the present invention, both the data value correlation and the trend correlation are positively correlated with the autoregressive coefficients. In one embodiment of the invention, after multiplying the data value correlation and the change trend correlation, an autoregressive coefficient is obtained, and because the value ranges of the data value correlation and the change trend correlation are all between 0 and one, the value range of the obtained autoregressive coefficient is also between 0 and 1, so that normalization of the autoregressive coefficient is realized. It should be noted that, in other embodiments of the present invention, the positive correlation mapping is normalized through other basic mathematical operations, which are not described and limited herein.
Further obtaining an error coefficient in the integrated moving average autoregressive model, wherein the error coefficient is considered as a deviation of a predicted value caused under the influence of other factors except modeling factors, and the error coefficient is needed to adjust the deviation so as to improve the prediction effect, so that the error coefficient can be obtained through the distribution difference and the numerical value difference between the predicted data and the actual data in the modeling stage.
Preferably, in one embodiment of the present invention, the method for obtaining an error coefficient includes:
setting an initial error coefficient to 0, and predicting data according to an integrated moving average autoregressive model corresponding to the initial error coefficient to obtain predicted data in a preset time range and a predicted data curve corresponding to the predicted data; obtaining actual data in a preset time range and a corresponding actual data curve thereof.
Obtaining the fluctuation characteristic difference of the data value between the predicted data curve and the actual data curve, wherein the larger the characteristic difference of the data value wave band is, the larger the characteristic difference of the data distribution between the predicted data and the actual data is, and the larger the explanation error is; obtaining a first average data value difference between the predicted data and the actual data, wherein the larger the first average data value difference is, the larger the deviation between the predicted data and the actual data is, and the larger the error coefficient is; and obtaining an error coefficient according to the data value fluctuation characteristic difference and the first average data value difference, wherein the data value fluctuation characteristic difference and the first average data value difference are in positive correlation with the error coefficient. The error coefficients are formulated in one embodiment of the invention as:
wherein,,is error coefficient +.>Representing the variance of the data values on the predicted data curve, i.e. representing the characteristic of the fluctuation of the data values by variance +.>Variance of data values on the actual data curve, +.>For the number of data that can be predicted within a predetermined time range, < >>Is->Predicted data->First->Actual data corresponding to the predicted data. It should be noted that since no history data exists before the initial time, no predicted data exists at the initial time, i.e. +.>Is->。
In the error coefficient formula, the variance is used for representing the fluctuation characteristic of the data value, and the difference of the fluctuation characteristic of the data value and the difference of the first average data value are combined in a product form to realize positive correlation mapping, so that an error coefficient is obtained. In other embodiments, the data value fluctuation feature may be represented by basic mathematical features such as standard deviation and fitting difference, and the positive correlation relationship may be represented by other mathematical operations, which are not limited and described herein.
Step S3: predicting predicted pressure regulating parameter data at a future moment by utilizing an integrated moving average autoregressive model according to the real-time pressure regulating parameter data; obtaining an adjustment coefficient according to the variation trend difference and the data difference between the predicted value and the actual data at each moment in the preset adjacent time range before the corresponding moment of the predicted voltage-regulating parameter data; and adjusting the predicted voltage regulation parameter data according to the adjustment coefficient to obtain the optimized voltage regulation parameter data.
And obtaining a corresponding integrated moving average autoregressive model according to the data in the corresponding preset time range at the real-time moment, and further obtaining predicted pressure regulating parameter data at the future moment according to the real-time pressure regulating parameter data. In order to further improve accuracy of the predicted data, an adjustment coefficient is obtained according to a change trend difference and a data difference between a predicted value and actual data at each moment in a preset adjacent time range before a corresponding moment of the predicted voltage-regulating parameter data, and because the predicted characteristic in the adjacent time range characterizes a predicted effect in the adjacent time range, that is, the change trend difference and the data difference between the predicted value and the actual data at each moment in the adjacent time range can characterize the predicted effect in the adjacent time range, the better the predicted effect is, the current model prediction capability is, the no adjustment is needed to be performed on the predicted voltage-regulating parameter data, the worse the predicted effect is, the certain error exists in the current model prediction capability, and further adjustment is needed to be performed on the predicted voltage-regulating parameter data.
Preferably, in one embodiment of the present invention, the method for obtaining the adjustment coefficient includes:
and acquiring second average data value differences between the predicted values and the actual data at all moments in the adjacent time range, namely characterizing the data differences through the second average data value differences.
Acquiring a third average tangential slope absolute value of all predicted values in a two-dimensional coordinate system at all moments in an adjacent time range; acquiring the absolute value of the fourth average tangential slope of all actual data in a two-dimensional coordinate system at all moments in the adjacent time range; and obtaining a second slope absolute value difference between the third average tangent slope absolute value and the fourth average tangent slope absolute value, namely representing the variation trend difference through the second slope absolute value difference.
And multiplying the second average data value difference by the second slope absolute value difference, and then carrying out normalization and negative correlation mapping to obtain an adjustment coefficient. In one embodiment of the invention the adjustment coefficient is formulated as:
wherein,,to adjust the coefficient +.>For the number of all moments in the vicinity of time, +.>Is->Predicted value at each moment, +.>Is->Actual value at each moment, +.>Absolute value of third average tangential slope for all predictors in adjacent time range, +.>Absolute value of the fourth average tangential slope, which is all the actual values in the adjacent time range, +.>Is a normalization function. In the embodiment of the invention, the normalization adopts the range normalization, and in other embodiments, other means can be adopted to perform normalization and negative correlation mapping, which are not limited and described in detail herein.
In the adjustment coefficient formula, the larger the product of the second average data value difference and the second slope absolute value difference is, the worse the prediction effect in the adjacent time period is, the smaller the adjustment coefficient obtained after normalization negative correlation processing is, and the more the prediction voltage-regulating parameter is required to be adjusted at the moment; on the contrary, the closer the processed adjustment coefficient is to 1, the more unnecessary adjustment of the predicted voltage regulation parameter is required at the moment.
Preferably, the adjustment coefficients in the predicted voltage regulation parameter data field are multiplied to obtain the optimized voltage regulation parameter data.
Step S4: the optimized pressure regulating parameter data in all dimensions form multidimensional optimized pressure regulating parameter data, and the pressure monitoring is carried out on the automatic gas regulating cabinet according to the multidimensional optimized pressure regulating parameter data.
The parameters of the automatic gas regulator cabinet in all dimensions are processed to obtain accurate prediction data in future time, namely, the multi-dimensional optimal pressure regulating parameter data is formed by the optimal pressure regulating parameter data in all dimensions. And (3) accurately monitoring the pressure of the automatic gas regulator according to the multidimensional optimized pressure regulating parameter data.
Preferably, after the monitoring of the pressure is achieved, the method further comprises: and calculating and normalizing the difference between the optimized voltage regulating parameter data in each dimension and the preset limit value in the corresponding dimension in the multidimensional optimized voltage regulating parameter data to obtain the state difference in each dimension. And carrying out pressure adjustment on the automatic gas regulator according to the relation between the average state difference under all dimensions and the preset judgment threshold value at real time. Further, in one embodiment of the invention, the state difference is characterized by the ratio between the specific data and the limit value, i.e. the closer the state difference is to 1, the worse the state within the pipeline is explained. Preferably, the judgment threshold value is set to be 0.85, and if the state difference is greater than 0.85, the condition of the pipeline is poor at the moment, and certain pressure relief operation is needed; and a second judgment threshold value of 0.3 can be further set, if the state difference is smaller than 0.3, certain pressurizing operation is needed, and the transmission of the fuel gas and the reasonable application of the fuel gas by a user are ensured. The pressure parameters at real time are adjusted by pre-analyzing the data at the future time, so that the safety of the pipeline is ensured in time. It should be noted that, because multidimensional data exists in the gas pipeline, a specific judgment threshold value can be set for the corresponding preset limit value data in each dimension, which is not described herein, for example, the specific judgment threshold value can be set according to the corresponding specific limit value data in a plurality of dimensions, such as the pipeline limit pressure, the pipeline limit gas rate, the pipeline limit concentration, and the like, which is not limited herein. In one embodiment of the present invention, further considering the multidimensional data feature, in order to prevent the contradiction between judging structures in different dimensions, the state difference in each dimension is averaged and then pressure control is further performed in combination with a judging threshold.
In summary, in the embodiment of the present invention, an integrated moving average autoregressive model is constructed through historical data, and autoregressive coefficients are constructed in the integrated moving average autoregressive model through data value correlation and variation trend correlation; and obtaining an error coefficient according to the distribution difference and the numerical value difference between the predicted data and the actual data. And further obtaining an adjustment coefficient according to the variation trend difference and the data difference between the predicted value and the actual value at each time in the adjacent time range corresponding to the obtained predicted pressure regulating parameter data, obtaining optimized pressure regulating parameter data according to the adjustment coefficient, and monitoring the pressure of the automatic gas regulator according to the optimized pressure regulating parameter data in multiple dimensions. According to the embodiment of the invention, the optimal pressure regulating parameter data with excellent accuracy is obtained by constructing the efficient integrated moving average autoregressive model and optimizing the predicted data, so that the pressure of the automatic gas regulator is monitored in time.
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 (8)
1. The utility model provides a gas automatic regulator cubicle pressure monitoring method based on multidimensional data which is characterized in that the method includes:
acquiring pressure regulating parameter data of the automatic gas pressure regulating cabinet at each moment in each dimension in a preset time range; the preamble data on the time sequence of the voltage regulation parameter data at each moment is the historical parameter data at the corresponding moment; the dimensions include at least gas pressure, gas velocity, and gas concentration;
constructing an integrated moving average autoregressive model according to the pressure regulating parameter data; the method for acquiring the autoregressive coefficients in the integrated moving average autoregressive model comprises the following steps: acquiring the data value correlation and the change trend correlation between the voltage regulation parameter data and the corresponding historical parameter data at each moment; obtaining an autoregressive coefficient at each moment according to the data value correlation and the change trend correlation; the error coefficient acquisition method in the integrated moving average autoregressive model comprises the following steps: obtaining the error coefficient according to the distribution difference and the numerical value difference between the predicted data and the actual data;
predicting predicted pressure regulating parameter data at a future moment by utilizing the integrated moving average autoregressive model according to the real-time pressure regulating parameter data; obtaining an adjustment coefficient according to the variation trend difference and the data difference between the predicted value and the actual data at each time in the preset adjacent time range before the corresponding time of the predicted voltage regulation parameter data; adjusting the predicted voltage regulation parameter data according to the adjustment coefficient to obtain optimized voltage regulation parameter data;
the optimized pressure regulating parameter data in all dimensions form multidimensional optimized pressure regulating parameter data, and the pressure monitoring is carried out on the automatic gas pressure regulating cabinet according to the multidimensional optimized pressure regulating parameter data;
the method for acquiring the change trend correlation comprises the following steps:
mapping the voltage regulating parameter data in the preset time range to a two-dimensional coordinate system, wherein the abscissa of the two-dimensional coordinate system is moment, and the ordinate is a data value; obtaining the absolute value of the tangential slope of a data point at each moment in the two-dimensional coordinate system;
acquiring a first average tangential slope absolute value of all data points of each voltage regulation parameter data within a preset judging time range; acquiring second average tangential slope absolute values corresponding to all data points in the preset time range;
obtaining the change trend correlation according to a first slope absolute value difference between the first average tangential slope absolute value and the second average tangential slope absolute value; the first slope absolute value difference and the change trend correlation are in a negative correlation relationship;
the method for acquiring the adjustment coefficient comprises the following steps:
acquiring second average data value differences between the predicted value and the actual data at all moments in the adjacent time range;
acquiring a third average tangential slope absolute value of all predicted values in the two-dimensional coordinate system at all moments in the adjacent time range; acquiring the absolute value of the fourth average tangential slope of all actual data in the two-dimensional coordinate system at all moments in the adjacent time range; acquiring a second slope absolute value difference between the third average tangential slope absolute value and the fourth average tangential slope absolute value;
and multiplying the second average data value difference by the second slope absolute value difference, and then carrying out normalization and negative correlation mapping to obtain the adjustment coefficient.
2. The method for monitoring the pressure of the automatic gas regulator cabinet based on multidimensional data according to claim 1, wherein the method for acquiring the correlation of the data values comprises the following steps:
acquiring a historical average data value of the historical parameter data corresponding to the voltage regulation parameter data at each moment, acquiring a first data value difference between the voltage regulation parameter data and the corresponding historical average data value, and acquiring the data value correlation according to the first data value difference, wherein the first data value difference and the data value correlation are in a negative correlation.
3. The method for monitoring the pressure of the automatic gas regulator cabinet based on multidimensional data according to claim 1, wherein the data value correlation and the change trend correlation are in positive correlation with the autoregressive coefficient.
4. The method for monitoring the pressure of the automatic gas regulator cubicle based on multidimensional data according to claim 1, wherein the method for acquiring the error coefficient comprises the following steps:
setting an initial error coefficient to 0, and predicting data according to an integrated moving average autoregressive model corresponding to the initial error coefficient to obtain predicted data and a predicted data curve corresponding to the predicted data in the preset time range; obtaining actual data in the preset time range and an actual data curve corresponding to the actual data;
obtaining a data value fluctuation characteristic difference between the predicted data curve and the actual data curve; obtaining a first average data value difference between the predicted data and the actual data; and obtaining the error coefficient according to the data value fluctuation characteristic difference and the first average data value difference, wherein the data value fluctuation characteristic difference and the first average data value difference are in positive correlation with the error coefficient.
5. The method for monitoring the pressure of the automatic gas regulator cubicle based on multidimensional data according to claim 1, wherein the method for acquiring the optimized pressure regulating parameter data comprises the following steps:
multiplying the predicted voltage regulation parameter data by the adjustment coefficient to obtain the optimized voltage regulation parameter data.
6. The method for monitoring the pressure of the automatic gas regulator cubicle based on the multidimensional data according to claim 1, wherein the method for monitoring the pressure of the automatic gas regulator cubicle based on the multidimensional optimized pressure regulating parameter data further comprises the following steps:
calculating and normalizing the difference between the optimized voltage regulating parameter data in each dimension and a preset limit value in the corresponding dimension in the multidimensional optimized voltage regulating parameter data to obtain a state difference in each dimension; and carrying out pressure adjustment on the automatic gas regulator according to the relation between the average state difference under all dimensions and the preset judgment threshold value at real time.
7. The method for monitoring the pressure of the automatic gas regulator cabinet based on the multidimensional data according to claim 6, wherein the judgment threshold is set to be 0.85, and when the state difference is larger than the judgment threshold, the automatic gas regulator cabinet is subjected to pressure adjustment at real time.
8. The method for monitoring the pressure of the automatic gas regulator cubicle based on multidimensional data according to claim 6, wherein the method for acquiring the state difference further comprises the following steps: and taking the ratio between the optimized voltage regulating parameter data in each dimension and the preset limit value in the corresponding dimension in the multidimensional optimized voltage regulating parameter data as the state difference.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310968144.5A CN116680661B (en) | 2023-08-03 | 2023-08-03 | Multi-dimensional data-based automatic gas regulator pressure monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310968144.5A CN116680661B (en) | 2023-08-03 | 2023-08-03 | Multi-dimensional data-based automatic gas regulator pressure monitoring method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116680661A CN116680661A (en) | 2023-09-01 |
CN116680661B true CN116680661B (en) | 2023-10-20 |
Family
ID=87785891
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310968144.5A Active CN116680661B (en) | 2023-08-03 | 2023-08-03 | Multi-dimensional data-based automatic gas regulator pressure monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116680661B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118070477A (en) * | 2023-11-28 | 2024-05-24 | 北京易动宇航科技有限公司 | Simulation credibility assessment method for electric propulsion system |
CN117536691B (en) * | 2024-01-09 | 2024-04-05 | 枣庄矿业集团新安煤业有限公司 | Fully-mechanized coal mining face equipment parameter monitoring method and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018045615A (en) * | 2016-09-16 | 2018-03-22 | 株式会社東芝 | Imbalance price prediction device, method, program and electric power trading system |
CN113359426A (en) * | 2021-07-09 | 2021-09-07 | 西安热工研究院有限公司 | On-line dynamic autoregressive prediction method based on boiler main steam pressure historical data |
RU213974U1 (en) * | 2022-04-26 | 2022-10-06 | Акционерное общество "Аэроприбор-Восход" | Air pressure receiver |
CN116383450A (en) * | 2023-06-05 | 2023-07-04 | 沧州中铁装备制造材料有限公司 | Railway and highway logistics transportation information comprehensive management system |
CN116448263A (en) * | 2023-06-16 | 2023-07-18 | 山东德圣源新材料有限公司 | Method for detecting running state of boehmite production equipment |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2139740C1 (en) * | 1996-07-23 | 1999-10-20 | Бражников Василий Степанович | Method of introduction into state of meditation |
-
2023
- 2023-08-03 CN CN202310968144.5A patent/CN116680661B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018045615A (en) * | 2016-09-16 | 2018-03-22 | 株式会社東芝 | Imbalance price prediction device, method, program and electric power trading system |
CN113359426A (en) * | 2021-07-09 | 2021-09-07 | 西安热工研究院有限公司 | On-line dynamic autoregressive prediction method based on boiler main steam pressure historical data |
RU213974U1 (en) * | 2022-04-26 | 2022-10-06 | Акционерное общество "Аэроприбор-Восход" | Air pressure receiver |
CN116383450A (en) * | 2023-06-05 | 2023-07-04 | 沧州中铁装备制造材料有限公司 | Railway and highway logistics transportation information comprehensive management system |
CN116448263A (en) * | 2023-06-16 | 2023-07-18 | 山东德圣源新材料有限公司 | Method for detecting running state of boehmite production equipment |
Non-Patent Citations (2)
Title |
---|
Variational Bayesian inference for network autoregression models;Wei-Ting Lai等;Computational Statistics & Data Analysis;第169卷;1-27 * |
预喷持续期对预燃室式天然气发动机性能的影响;李晓东等;内燃机与动力装置;第39卷(第05期);27-31 * |
Also Published As
Publication number | Publication date |
---|---|
CN116680661A (en) | 2023-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116680661B (en) | Multi-dimensional data-based automatic gas regulator pressure monitoring method | |
US20160116892A1 (en) | Method and system of cause analysis and correction for manufacturing data | |
CN115933787B (en) | Indoor multi-terminal intelligent control system based on indoor environment monitoring | |
CN116400126B (en) | Low-voltage power box with data processing system | |
CN116335925B (en) | Data enhancement-based intelligent regulation and control system for underground coal mine emulsification pump station | |
CN114997309A (en) | Water feed pump fault early warning method and device | |
CN111754029A (en) | Community load prediction system | |
CN116433085A (en) | Performance evaluation method of rolling process control system | |
US7155367B1 (en) | Method for evaluating relative efficiency of equipment | |
CN116629843B (en) | Remote early warning and maintenance decision support system of intelligent diesel generator set | |
CN111445141B (en) | Load distribution method, system and device of heat supply unit | |
CN112066264A (en) | Pipeline hydraulic pressure regulation and control method, device and system | |
CN112232580B (en) | Power supply interruption loss analysis method and device | |
CN115450710A (en) | Method for optimizing sliding pressure operation of steam turbine | |
CN118194607B (en) | Dynamic modeling method and system for boiler unit data | |
CN118094446B (en) | Anaerobic system running condition intelligent analysis method based on machine learning | |
CN118194607A (en) | Dynamic modeling method and system for boiler unit data | |
CN117927998B (en) | Mining area heating system intelligent regulation and control method based on data analysis | |
CN115114983B (en) | Method for acquiring and analyzing electric quantity data based on big data equipment and computer system | |
CN117871771B (en) | Big data-based gas energy monitoring method | |
CN115600930B (en) | Intelligent management method for electric power fitting production data | |
CN117828371B (en) | Intelligent analysis method for business information of comprehensive operation and maintenance platform | |
CN117584790B (en) | Capacity-free charging pile control system | |
CN117032017A (en) | Automatic frequency conversion control system of intelligent pressure control non-negative pressure water supply equipment | |
CN117391391A (en) | Low-carbon energy consumption optimization system and method based on carbon emission double control |
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 | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A Pressure Monitoring Method for Gas Automatic Pressure Regulating Cabinet Based on Multidimensional Data Granted publication date: 20231020 Pledgee: Changsha Lixiang Road Sub Branch of China Everbright Bank Co.,Ltd. Pledgor: HUNAN TIANLIAN CITY DATA CONTROL Co.,Ltd. Registration number: Y2024980019060 |