CN114819065A - Gas concentration prediction method and system for optimizing LSTM based on cuckoo search algorithm - Google Patents
Gas concentration prediction method and system for optimizing LSTM based on cuckoo search algorithm Download PDFInfo
- Publication number
- CN114819065A CN114819065A CN202210358187.7A CN202210358187A CN114819065A CN 114819065 A CN114819065 A CN 114819065A CN 202210358187 A CN202210358187 A CN 202210358187A CN 114819065 A CN114819065 A CN 114819065A
- Authority
- CN
- China
- Prior art keywords
- gas concentration
- prediction model
- lstm
- lstm prediction
- data
- 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.)
- Granted
Links
- 241000544061 Cuculus canorus Species 0.000 title claims abstract description 48
- 238000010845 search algorithm Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000003245 coal Substances 0.000 claims abstract description 74
- 238000011156 evaluation Methods 0.000 claims abstract description 54
- 238000005457 optimization Methods 0.000 claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 6
- 210000002569 neuron Anatomy 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 241001441723 Takifugu Species 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Combustion & Propulsion (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a method and a system for predicting gas concentration by optimizing LSTM based on a cuckoo search algorithm. The method comprises the following steps: processing the acquired coal mine underground gas concentration data to obtain standardized gas concentration data; determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model; constructing a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimization parameter combination of the LSTM prediction model based on a cuckoo search algorithm; and according to the monitoring data of the gas concentration under the coal mine, carrying out accuracy evaluation on the CS-LSTM prediction model. Therefore, the CS-LSTM prediction model which is optimized is used for predicting the coal mine gas concentration, the accuracy of coal mine gas concentration prediction is improved, and coal mine safety production is guided.
Description
Technical Field
The application relates to the technical field of gas concentration prediction, in particular to a method and a system for predicting gas concentration based on optimization of LSTM by a cuckoo search algorithm.
Background
Coal mine gas disasters are one of major disasters threatening the safety production of coal mines, the gas disasters directly obstruct the normal production of the coal mines, the sustainable, stable and healthy development of the coal industry is obstructed, and the strengthening of gas disaster prevention and control is an important guarantee for ensuring the stable and reliable supply of coal energy and promoting the comprehensive and healthy development of national economy.
At present, with the emphasis of the country on coal mine safety production and the requirement of development of coal mine enterprises per se, the method has important significance for accurately predicting the gas concentration and forecasting and preventing gas outburst disasters. The traditional gas prediction method is used for predicting whether one or more indexes exceed a critical value according to quantitative indexes of the properties and occurrence conditions of the gas-containing coal body, such as coal seam property indexes, gas indexes, ground stress indexes or comprehensive indexes, but the prediction precision of the traditional prediction method cannot meet the requirement of coal mine safety production.
Therefore, there is a need to provide an improved solution to the above-mentioned deficiencies of the prior art.
Disclosure of Invention
The application aims to provide a method and a system for predicting gas concentration based on an LSTM optimized by a cuckoo search algorithm, so as to solve or alleviate the problems in the prior art.
In order to achieve the above purpose, the present application provides the following technical solutions:
the application provides a cuckoo search algorithm-based LSTM gas concentration prediction method, which comprises the following steps: step S101, processing the acquired coal mine underground gas concentration data to obtain standardized gas concentration data; s102, determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model; s103, constructing a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimized parameter combination of the LSTM prediction model based on a cuckoo search algorithm; and S104, evaluating the accuracy of the CS-LSTM prediction model according to the underground gas concentration monitoring data of the coal mine.
Preferably, in step S101, the processing the acquired coal mine underground gas concentration data to obtain standardized gas concentration data includes: interpolating missing data in the gas concentration data based on a spline interpolation method; according to the interpolated gas concentration data, according to the formula:
obtaining standardized gas concentration data; wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
Preferably, the interpolating missing data in the gas concentration data based on the spline interpolation method includes: on the basis of a spline interpolation method, according to the acquired coal mine underground gas concentration data and a preset interpolation model, carrying out interpolation on missing data in the gas concentration data; the preset interpolation model is as follows:
wherein x is 1 ，x 2 Two adjacent data points, x, before the missing value in the acquired coal mine underground gas concentration data are represented 3 ，x 4 Representing missing values in the acquired coal mine underground gas concentration dataThe last two adjacent data points, and x 1 ，x 2 ，x 3 ，x 4 All satisfy f (x) ═ a i x 2 +b i x+c i ，i＝1，2，3；x＝(x 1 ，x 2 ，x 3 ，x 4 )。
Preferably, in step S102, determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model includes: determining the LSTM layer number and the corresponding neuron number of the LSTM prediction model, and the full-connection layer number and the corresponding neuron number of the LSTM prediction model as an optimized parameter combination of the LSTM prediction model; and determining the mean square error as an accuracy evaluation index of the LSTM prediction model.
Preferably, in step S103, based on the cuckoo search algorithm, the CS-LSTM prediction model is constructed according to the accuracy evaluation index of the LSTM prediction model and the optimized parameter combination of the LSTM prediction model, including: calculating the reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm, and inputting the optimized parameter combination of the LSTM prediction model corresponding to the maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model.
Preferably, in step S104, the evaluating the accuracy of the CS-LSTM prediction model according to the monitoring data of the gas concentration in the underground coal mine includes: according to the formula:
calculating the average absolute error MAE and the root mean square error RMSE of the gas concentration under the coal mine, and carrying out accuracy evaluation on the CS-LSTM prediction model; wherein n represents the tileThe number of concentration monitoring data; x is the number of j Monitoring data of the gas concentration under the coal mine; x is the number of j ' represents the predicted value of the gas concentration obtained by the CS-LSTM prediction model.
The embodiment of the present application further provides a gas concentration prediction system based on an LSTM optimized by a cuckoo search algorithm, including: the preprocessing unit is configured to process the acquired coal mine underground gas concentration data to obtain standardized gas concentration data; the parameter index unit is configured to determine an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model; the optimization unit is configured to construct a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimization parameter combination of the LSTM prediction model based on a cuckoo search algorithm; and the evaluation unit is configured to evaluate the accuracy of the CS-LSTM prediction model according to the gas concentration monitoring data in the coal mine.
Preferably, the preprocessing unit includes an interpolation subunit, configured to interpolate missing data in the gas concentration data based on a spline difference method; a normalization subunit configured to, based on the interpolated gas concentration data, according to a formula:
obtaining standardized gas concentration data; wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
Preferably, the optimization unit is further configured to calculate a reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm, and input the optimized parameter combination of the LSTM prediction model corresponding to the maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model.
Preferably, the evaluation unit is further configured to:
calculating the average absolute error MAE and the root mean square error RMSE of the gas concentration under the coal mine, and carrying out accuracy evaluation on the CS-LSTM prediction model; wherein n represents the number of the gas concentration detection data; x is the number of j Detecting data of the gas concentration under the coal mine; x is the number of j ' represents the predicted value of the gas concentration obtained by the CS-LSTM prediction model.
Has the advantages that:
according to the technical scheme, the method comprises the steps that acquired coal mine underground gas concentration data are processed to obtain standardized gas concentration data; determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model; then, based on a cuckoo search algorithm, according to the accuracy evaluation index of the LSTM prediction model and the optimization parameter combination of the LSTM prediction model, constructing to obtain a CS-LSTM prediction model; and finally, evaluating the accuracy of the obtained CS-LSTM prediction model according to the underground gas concentration monitoring data of the coal mine. Therefore, the CS-LSTM prediction model which is optimized is used for predicting the coal mine gas concentration, the accuracy of coal mine gas concentration prediction is improved, and coal mine safety production is guided.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. Wherein:
fig. 1 is a schematic flow chart of a gas concentration prediction method for optimizing LSTM based on cuckoo search algorithm according to some embodiments of the present application;
FIG. 2 is a technical logic diagram of a gas concentration prediction method for optimizing LSTM based on a cuckoo search algorithm according to some embodiments of the present application;
fig. 3 is a schematic structural diagram of a gas concentration prediction system for optimizing LSTM based on cuckoo search algorithm according to some embodiments of the present application.
Detailed Description
The present application will be described in detail below with reference to the embodiments with reference to the attached drawings. The various examples are provided by way of explanation of the application and are not limiting of the application. In fact, it will be apparent to those skilled in the art that modifications and variations can be made in the present application without departing from the scope or spirit of the application. For instance, features illustrated or described as part of one embodiment, can be used with another embodiment to yield a still further embodiment. It is therefore intended that the present application cover the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents.
In the description of the present application, the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description of the present application but do not require that the present application must be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present application. The terms "connected," "connected," and "disposed" as used herein are intended to be broadly construed, and may include, for example, fixed and removable connections; can be directly connected or indirectly connected through intermediate components; the specific meaning of the above terms can be understood as the case may be, to one of ordinary skill in the art.
Because the traditional coal mine gas concentration prediction is to predict whether one or more indexes exceed a critical value or not according to coal bed property indexes, gas indexes, ground stress indexes or comprehensive indexes, which are quantitative indexes of the properties and occurrence conditions of the gas-containing coal, the prediction precision is difficult to meet the requirement of coal mine safe production. For this reason, the applicant proposes the gas concentration prediction scheme based on the cuckoo search algorithm optimization LSTM, optimizes the hyper-parameters of the LSTM prediction model by constructing a basic frame of the LSTM prediction model and utilizing the cuckoo optimization algorithm, and endows the optimized result to the final LSTM prediction model to obtain a CS-LSTM prediction model; and then, the accuracy of the CS-LSTM prediction model is evaluated by using actual gas concentration detection data and evaluation indexes, so that the CS-LSTM prediction model can accurately predict the coal mine gas concentration, and the prediction accuracy of the coal mine gas concentration is improved.
Exemplary method
Fig. 1 is a schematic flow chart of a gas concentration prediction method for optimizing LSTM based on cuckoo search algorithm according to some embodiments of the present application; FIG. 2 is a technical logic diagram of a gas concentration prediction method for optimizing LSTM based on a cuckoo search algorithm according to some embodiments of the present application; as shown in fig. 1 and 2, the prediction method for optimizing LSTM gas concentration based on cuckoo search algorithm includes:
step S101, processing the acquired coal mine underground gas concentration data to obtain standardized gas concentration data;
in the embodiment of the application, the underground gas concentration data of the coal mine is acquired by the underground gas concentration sensor and transmitted to the ground monitoring center through the underground communication network. Because the gas concentration sensor positioned under the coal mine is affected by the adverse effect of the surrounding environment, the acquired data can have noise, loss and the like inevitably, so that in the embodiment of the application, the acquired gas concentration data is preprocessed and standardized, abnormal data and missing data in the acquired gas concentration data under the coal mine are corrected, and the gas concentration data is standardized.
Specifically, when the acquired coal mine underground gas concentration data is processed to obtain standardized gas concentration data, firstly, missing data in the gas concentration data is interpolated based on a spline interpolation method; then, the normalized gas concentration data is obtained according to the formula (1) based on the interpolated gas concentration data. Equation (1) is as follows:
wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
In the embodiment of the present application, missing data in the gas concentration data is interpolated based on a spline interpolation method, specifically: and interpolating missing data in the gas concentration data according to the acquired underground coal mine gas concentration data and a preset interpolation model based on a spline interpolation method. The preset interpolation model is shown as formula (2), and the formula (2) is as follows:
in the formula, x 1 ，x 2 Two adjacent data points, x, before the missing value in the acquired coal mine underground gas concentration data are represented 3 ，x 4 Two adjacent data points after the missing value in the obtained coal mine startle gas concentration data are shown, and x 1 ，x 2 ，x 3 ，x 4 All satisfy f (x) ═ a i x 2 +b i x+c i ，i＝1，2，3；x＝(x 1 ，x 2 ，x 3 ，x 4 )。
In the embodiment of the present application, first, two adjacent data points are taken before and after the missing value, respectively, and the missing value is ensured to be located between the preceding and following value-taking points, and each value-taking point satisfies f (x) a i x 2 +b i x+c i (ii) a And then, constructing a preset interpolation model based on a preset principle. Wherein, the preset principle comprises: f (x) ═ a i x 2 +b i x+c i The values at the nodes are equal, i.e. f (x) a i x 2 +b i x+c i At x 1 ，x 3 The values at the two points are equal; the first endpoint and the last endpoint must passAn equation (i.e., f (x) ═ a) 1 x 2 +b 1 x+c 1 ) And the last equation (i.e., f (x) ═ a) 4 x 2 +b 4 x+c 4 )；f(x)＝a i x 2 +b i x+c i At node (x) 1 ，x 3 ) The values of the first derivatives at (a) are equal; first equation f (x) ═ a 1 x 2 +b 1 x+c 1 Is zero, i.e. a 1 0. Then, the formula (2) is solved to obtain a function f (x) of the corresponding section, and the missing value is brought into the function f (x) of the corresponding section to obtain the missing value. And after the missing value interpolation is finished, changing the interpolated gas concentration data according to a formula (1) to obtain standardized gas concentration data.
S102, determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model;
in the embodiment of the application, in the reiteration process of the LSTM prediction model, parameters such as each weight coefficient and the like can be automatically updated to find an optimal value, and the hyper-parameter is directly set before the iteration of the LSTM prediction model starts and cannot be updated along with the iteration; therefore, the LSTM layer number and the corresponding neuron number of the LSTM prediction model, the full-connection layer number and the corresponding neuron number of the LSTM prediction model are optimized parameter combinations of the LSTM prediction model, namely four hyper-parameters of the LSTM layer number and the corresponding neuron number, the full-connection layer number and the corresponding neuron number are set as parameter combinations needing to be optimized. Therefore, the prediction accuracy of the network result of the prediction model on the coal mine underground gas concentration data is effectively improved. The mean square error is used as an accuracy evaluation index of the LSTM prediction model, and the smaller the mean square error is, the higher the prediction accuracy of the gas concentration data by the prediction model is.
And S103, constructing a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm.
In the embodiment of the application, the LSTM prediction model is optimized through a cuckoo search algorithm, specifically, a hyper-parameter basic frame of the LSTM prediction model is input into the cuckoo search algorithm as an optimization object, the inverse of the mean square error is used as a moderate function of the cuckoo search algorithm, and the maximum value of the moderate function is the optimal value of cuckoo search.
Specifically, based on a cuckoo search algorithm, calculating the reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model, and inputting the optimized parameter combination of the LSTM prediction model corresponding to the maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model. Namely, the optimum value of cuckoo search corresponding to the maximum value of the moderate function is used as the optimum hyper-parameter combination of the LSTM prediction model and is input into the LSTM prediction model, and the CS-LSTM prediction model can be obtained.
And S104, evaluating the accuracy of the CS-LSTM prediction model according to the underground gas concentration monitoring data of the coal mine.
In the embodiment of the application, the accuracy of the CS-LSTM prediction model is evaluated by using the actually monitored coal mine underground gas concentration monitoring data and the determined evaluation index. Specifically, according to the formula:
and calculating the average absolute error MAE and the root mean square error RMSE of the gas concentration of the frightened coal mine, and carrying out accuracy evaluation on the CS-LSTM prediction model. Wherein n represents the number of gas concentration monitoring data; x is the number of j Monitoring data of the underground gas concentration of the coal mine; x is the number of j ' expressing the gas concentration prediction value obtained by CS-LSTM prediction model
In the embodiment of the application, the CS-LSTM prediction model is obtained by constructing a basic framework of the LSTM prediction model, optimizing the hyper-parameters of the LSTM prediction model by using a cuckoo optimization algorithm and endowing the optimizing result to the final LSTM prediction model; and then, the accuracy of the CS-LSTM prediction model is evaluated by using actual gas concentration detection data and evaluation indexes, so that the CS-LSTM prediction model can accurately predict the coal mine gas concentration, and the prediction accuracy of the coal mine gas concentration is improved.
Exemplary System
FIG. 3 is a schematic diagram of a gas concentration prediction system for optimizing LSTM based on a cuckoo search algorithm according to some embodiments of the present application; as shown in fig. 3, the gas concentration prediction system based on the cuckoo search algorithm to optimize LSTM includes: the device comprises a preprocessing unit, a parameter index unit, an optimization unit and an evaluation unit. The preprocessing unit is configured to process the acquired coal mine underground gas concentration data to obtain standardized gas concentration data; the parameter index unit is configured to determine an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model; the optimization unit is configured to construct a CS-LSTM prediction model based on a cuckoo search algorithm according to the accuracy evaluation index of the LSTM prediction model and the optimization parameter combination of the LSTM prediction model; the evaluation unit is configured to perform accuracy evaluation on the CS-LSTM prediction model according to the gas concentration monitoring data in the coal mine.
In some alternative embodiments, the pre-processing unit includes an interpolation subunit and a normalization subunit. The interpolation subunit is configured to interpolate missing data in the gas concentration data based on a spline difference method; the normalization subunit is configured to, based on the interpolated gas concentration data, according to the formula:
obtaining standardized gas concentration data; wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
Further, the interpolation subunit is further configured to interpolate missing data in the gas concentration data according to a preset interpolation model and the acquired coal mine underground gas concentration data based on a spline interpolation method; the preset interpolation model is as follows:
wherein x is 1 ，x 2 Two adjacent data points, x, before the missing value in the acquired coal mine underground gas concentration data are represented 3 ，x 4 Representing two adjacent data points after the missing value in the acquired coal mine underground gas concentration data, and x 1 ，x 2 ，x 3 ，x 4 All satisfy f (x) ═ a i x 2 +b i x+c i ，i＝1，2，3；x＝(x 1 ，x 2 ，x 3 ，x 4 )。
In some optional embodiments, the parameter indicator unit is further configured to determine that the number of LSTM layer numbers and their corresponding neuron numbers, the number of fully-connected layer numbers and their corresponding neuron numbers of the LSTM prediction model are an optimized parameter combination of the LSTM prediction model; and determining the mean square error as an accuracy evaluation index of the LSTM prediction model.
In some optional embodiments, the optimization unit is further configured to calculate a reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm, and input the optimized parameter combination of the LSTM prediction model corresponding to a maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model.
In some optional embodiments, the evaluation unit is further configured to:
underground coal mine calculationThe average absolute error MAE and the root mean square error RMSE of the gas concentration, and the accuracy evaluation is carried out on the CS-LSTM prediction model; wherein n represents the number of gas concentration detection data; x is the number of j Detecting data of the underground gas concentration of the coal mine; x is the number of j ' denotes the predicted gas concentration value obtained by the CS-LSTM prediction model.
The cuckoo search algorithm-based LSTM gas concentration prediction system provided by the embodiment of the application can realize the steps and the flow of any cuckoo search algorithm-based LSTM gas concentration prediction method, and achieves the same technical effects, which are not described in detail herein.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. A gas concentration prediction method for optimizing LSTM based on cuckoo search algorithm is characterized by comprising the following steps:
step S101, processing the acquired coal mine underground gas concentration data to obtain standardized gas concentration data;
s102, determining an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model;
s103, constructing a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimized parameter combination of the LSTM prediction model based on a cuckoo search algorithm;
and S104, evaluating the accuracy of the CS-LSTM prediction model according to the underground gas concentration monitoring data of the coal mine.
2. The method for predicting gas concentration based on optimized LSTM by the cuckoo search algorithm as claimed in claim 1, wherein in step S101, the processing of the acquired coal mine underground gas concentration data to obtain standardized gas concentration data comprises:
interpolating missing data in the gas concentration data based on a spline interpolation method;
according to the interpolated gas concentration data, according to the formula:
obtaining standardized gas concentration data;
wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
3. The gas concentration prediction method for optimizing LSTM by the Fugu bird search algorithm as claimed in claim 2, wherein the interpolation of missing data in the gas concentration data based on the spline interpolation method comprises:
on the basis of a spline interpolation method, according to the acquired coal mine underground gas concentration data and a preset interpolation model, carrying out interpolation on missing data in the gas concentration data;
the preset interpolation model is as follows:
wherein x is 1 ，x 2 Two adjacent data points, x, before the missing value in the acquired coal mine underground gas concentration data are represented 3 ，x 4 Representing two adjacent data points after the missing value in the acquired coal mine underground gas concentration data, and x 1 ，x 2 ，x 3 ，x 4 All satisfy f (x) ═ a i x 2 +b i x+c i ，i＝1，2，3；x＝(x 1 ，x 2 ，x 3 ，x 4 )。
4. The method for predicting gas concentration based on optimized LSTM by the cuckoo search algorithm as claimed in claim 1, wherein in step S102, determining the optimized parameter combination of the pre-constructed LSTM prediction model and the accuracy evaluation index of the LSTM prediction model comprises:
determining the LSTM layer number and the corresponding neuron number of the LSTM prediction model, and the full-connection layer number and the corresponding neuron number of the LSTM prediction model as an optimized parameter combination of the LSTM prediction model;
and determining the mean square error as an accuracy evaluation index of the LSTM prediction model.
5. The method for predicting gas concentration based on cuckoo search algorithm optimization LSTM of claim 1, wherein in step S103, based on cuckoo search algorithm, according to the accuracy evaluation index of the LSTM prediction model and the optimized parameter combination of the LSTM prediction model, constructing a CS-LSTM prediction model, comprises:
calculating the reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm, and inputting the optimized parameter combination of the LSTM prediction model corresponding to the maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model.
6. The method for predicting gas concentration based on optimized LSTM by the cuckoo search algorithm as claimed in claim 1, wherein in step S104, the evaluating accuracy of the CS-LSTM prediction model according to the monitoring data of gas concentration in the underground coal mine comprises:
according to the formula:
calculating the average absolute error MAE and the root mean square error RMSE of the gas concentration under the coal mine, and carrying out accuracy evaluation on the CS-LSTM prediction model;
wherein n represents the number of the gas concentration monitoring data; x is the number of j Monitoring data of the gas concentration under the coal mine; x is the number of j ' represents the predicted value of the gas concentration obtained by the CS-LSTM prediction model.
7. A gas concentration prediction system based on a cuckoo search algorithm for optimizing LSTM is characterized by comprising:
the preprocessing unit is configured to process the acquired coal mine underground gas concentration data to obtain standardized gas concentration data;
the parameter index unit is configured to determine an optimized parameter combination of a pre-constructed LSTM prediction model and an accuracy evaluation index of the LSTM prediction model;
the optimization unit is configured to construct a CS-LSTM prediction model according to the accuracy evaluation index of the LSTM prediction model and the optimization parameter combination of the LSTM prediction model based on a cuckoo search algorithm;
and the evaluation unit is configured to evaluate the accuracy of the CS-LSTM prediction model according to the gas concentration monitoring data in the coal mine.
8. The method for predicting gas concentration based on cuckoo search algorithm optimization LSTM according to claim 7, wherein the preprocessing unit comprises:
the interpolation subunit is configured to interpolate missing data in the gas concentration data based on a spline difference method;
a normalization subunit configured to, based on the interpolated gas concentration data, according to a formula:
obtaining standardized gas concentration data;
wherein x is * The normalized gas concentration data is shown, and f (x) shows the interpolated gas concentration data.
9. The method as claimed in claim 7, wherein the optimizing unit is further configured to calculate a reciprocal of the accuracy evaluation index of the LSTM prediction model according to the optimized parameter combination of the LSTM prediction model based on the cuckoo search algorithm, and input the optimized parameter combination of the LSTM prediction model corresponding to a maximum value of the reciprocal of the accuracy evaluation index of the LSTM prediction model into the LSTM prediction model to obtain the CS-LSTM prediction model.
10. The method of claim 7, wherein the evaluation unit is further configured to respectively calculate the gas concentration according to the formula:
calculating the average absolute error MAE and the root mean square error RMSE of the gas concentration under the coal mine, and carrying out accuracy evaluation on the CS-LSTM prediction model;
wherein n represents the number of the gas concentration detection data; x is the number of j Detecting data of the gas concentration under the coal mine; x is the number of j ' represents the predicted value of the gas concentration obtained by the CS-LSTM prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210358187.7A CN114819065B (en) | 2022-04-06 | 2022-04-06 | Method and system for predicting gas concentration by optimizing LSTM (least squares) based on cuckoo search algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210358187.7A CN114819065B (en) | 2022-04-06 | 2022-04-06 | Method and system for predicting gas concentration by optimizing LSTM (least squares) based on cuckoo search algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114819065A true CN114819065A (en) | 2022-07-29 |
| CN114819065B CN114819065B (en) | 2024-04-05 |
Family
ID=82535602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210358187.7A Active CN114819065B (en) | 2022-04-06 | 2022-04-06 | Method and system for predicting gas concentration by optimizing LSTM (least squares) based on cuckoo search algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114819065B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970182A (en) * | 2024-03-28 | 2024-05-03 | 国网山东省电力公司曲阜市供电公司 | Electric leakage early warning method and system based on DTW algorithm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200110998A1 (en) * | 2018-10-09 | 2020-04-09 | Ebay Inc. | System and method for improving user engagement based on user session analysis |
| CN112730796A (en) * | 2020-12-14 | 2021-04-30 | 重庆大学 | Coal and gas outburst risk evaluation method |
| CN113190535A (en) * | 2021-01-19 | 2021-07-30 | 云南卫士盾科技有限公司 | Gas concentration prediction method based on improved grey-long and short-term memory neural network combined model |
| US20210390230A1 (en) * | 2020-06-16 | 2021-12-16 | Chongqing University | Method for Quickly Optimizing Key Mining Parameters of Outburst Coal Seam |
| CN114169594A (en) * | 2021-11-25 | 2022-03-11 | 安徽理工大学 | Gas concentration prediction method based on LSTM-LightGBM variable weight combined model |
-
2022
- 2022-04-06 CN CN202210358187.7A patent/CN114819065B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200110998A1 (en) * | 2018-10-09 | 2020-04-09 | Ebay Inc. | System and method for improving user engagement based on user session analysis |
| US20210390230A1 (en) * | 2020-06-16 | 2021-12-16 | Chongqing University | Method for Quickly Optimizing Key Mining Parameters of Outburst Coal Seam |
| CN112730796A (en) * | 2020-12-14 | 2021-04-30 | 重庆大学 | Coal and gas outburst risk evaluation method |
| CN113190535A (en) * | 2021-01-19 | 2021-07-30 | 云南卫士盾科技有限公司 | Gas concentration prediction method based on improved grey-long and short-term memory neural network combined model |
| CN114169594A (en) * | 2021-11-25 | 2022-03-11 | 安徽理工大学 | Gas concentration prediction method based on LSTM-LightGBM variable weight combined model |
Non-Patent Citations (1)
| Title |
|---|
| 梁运培: "基于CS-LSTM的工作面瓦斯浓度智能预测研究", 矿业安全与环保, vol. 49, no. 04, 22 July 2022 (2022-07-22), pages 80 - 86 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970182A (en) * | 2024-03-28 | 2024-05-03 | 国网山东省电力公司曲阜市供电公司 | Electric leakage early warning method and system based on DTW algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114819065B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Critical review of data-driven decision-making in bridge operation and maintenance | |
| CN109409567B (en) | Complex equipment residual life prediction method based on double-layer long-short term memory network | |
| CN114819065A (en) | Gas concentration prediction method and system for optimizing LSTM based on cuckoo search algorithm | |
| CN107862763B (en) | Train safety early warning evaluation model training method, module and monitoring evaluation system | |
| CN112101636A (en) | GRU and GARCH-based satellite long-period variance-variance degradation prediction evaluation method | |
| AU2020102863A4 (en) | Digital-twin-driven fault prognosis method and system for subsea production system of offshore oil | |
| CN114969953B (en) | Optimized shield underpass tunnel design method and equipment based on Catboost-NSGA-III | |
| CN110457786B (en) | Ship unloader association rule fault prediction model method based on deep confidence network | |
| Zhang et al. | Research and application of improved gas concentration prediction model based on grey theory and BP neural network in digital mine | |
| CN103793854A (en) | Multiple combination optimization overhead transmission line operation risk informatization assessment method | |
| CN102332011A (en) | Method for selecting effective data of in-orbit spacecraft | |
| CN108873859A (en) | Based on the bridge-type grab ship unloader fault prediction model method for improving correlation rule | |
| CN111091241A (en) | BP neural network-based drug sales prediction and decision method and system | |
| CN113987908A (en) | Natural gas pipe network leakage early warning method based on machine learning method | |
| CN114925718A (en) | Transformer core grounding current online prediction method and device, storage medium and fault diagnosis method | |
| CN113408190B (en) | Surrounding rock deformation prediction method for highway tunnel construction period based on Bayes-LSTM model | |
| CN117094180A (en) | Digital twin model quantization manufacturing system | |
| CN108062639A (en) | A kind of Risk Propagation Model and the algorithm suitable for the model | |
| CN110827134A (en) | Power grid enterprise financial health diagnosis method | |
| CN108108837A (en) | A kind of area new energy power supply structure optimization Forecasting Methodology and system | |
| CN110990938A (en) | Soft measurement method for rock digging state of tunnel face | |
| Zhou et al. | Quantification of value of information associated with optimal observation actions within partially observable Markov decision processes | |
| CN106529725A (en) | Gas outburst prediction method based on firefly algorithm and SOM network | |
| CN115099129A (en) | Natural gas well yield prediction method based on input characteristic error correction | |
| CN113705923A (en) | Load prediction method based on intelligent algorithm |
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 |