CN111310317A - Granary space-time temperature field prediction method and device based on big data and interpolation prediction - Google Patents
Granary space-time temperature field prediction method and device based on big data and interpolation prediction Download PDFInfo
- Publication number
- CN111310317A CN111310317A CN202010079018.0A CN202010079018A CN111310317A CN 111310317 A CN111310317 A CN 111310317A CN 202010079018 A CN202010079018 A CN 202010079018A CN 111310317 A CN111310317 A CN 111310317A
- Authority
- CN
- China
- Prior art keywords
- temperature
- data
- layer
- granary
- training
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a granary space-time temperature field prediction method and device based on big data and interpolation prediction. The method comprises the following steps: reading grain condition data and warehouse information; intercepting different time point data as data basis; 3D rebuilding the temperature in the bin; selecting a section to be predicted, and predicting the temperature time sequence of the section; judging whether the coordinate position is an outer layer coordinate point or not for the temperature point to be predicted in the selected section; inputting sample data into an input sequence; initializing, and setting a training period and precision; calculating the actual output of the network layer; calculating errors and correcting the weight and the threshold; training a neural network and judging whether the training precision reaches a preset precision or whether the training period reaches a preset period, if so, predicting a temperature value; and obtaining a predicted temperature field diagram of the cross section by using the predicted cross section discrete temperature point space interpolation. The method accurately predicts the change trend of the temperature field in the granary, intuitively and effectively reflects the internal temperature information of the granary and grain pile, and ensures the grain storage quality.
Description
Technical Field
The invention relates to a temperature field prediction method in the technical field of grain storage, in particular to a granary space-time temperature field prediction method based on big data and interpolation prediction, and further relates to a granary space-time temperature field prediction device based on big data and interpolation prediction and applying the method.
Background
At present, a plurality of large-scale grain depots are built in China, the capacity of each single granary is multiple times of that of the single granary built in the past, and the problems of mildew and insect damage in the granaries are more serious. Grain is used as a special and complex life body, and the change rule of the temperature field inside the grain stack also becomes abnormally complex, so that the grain temperature distribution can be accurately mastered, and the change trend of the grain temperature can be reasonably analyzed and predicted, which is one of the important methods for pre-judging the grain storage safety state.
However, the current grain temperature prediction scheme in the granary aims at the temperature prediction of a single point, for a complex storage environment, the single-point temperature prediction is not enough to show the change trend of the temperature of the whole granary, the overall temperature distribution cannot be shown in the large-size granary environment, the situation that local temperature abnormality occurs in the granary cannot be predicted possibly, the deviation between theoretical prediction and actual situation is large, the change of the hot core and the cold core of the granary cannot be shown, and the false judgment occurs to a grain situation manager.
Disclosure of Invention
The invention provides a granary space-time temperature field prediction method and device based on big data and interpolation prediction, and aims to solve the technical problems that the existing granary grain temperature prediction scheme cannot predict the specific temperature distribution of the whole granary, and mathematical modeling analysis on each granary is not practical.
The invention is realized by adopting the following technical scheme: a granary space-time temperature field prediction method based on big data and interpolation prediction comprises the following steps:
step S1, reading grain situation data and corresponding warehouse information from a granary database;
step S2, intercepting different time point data according to the warehouse state in the warehouse information;
step S3, performing 3D space reconstruction on the temperature to obtain position information corresponding to the temperature point;
step S4, selecting a section to be predicted, and firstly performing time sequence prediction of the section temperature;
step S5, judging whether the temperature points of different positions of the cross section are outer layer temperature points or not;
when the temperature point of the location is the outer layer temperature point, performing step S6;
step S6, selecting variables including meteorological data, internal temperature and humidity and adjacent temperature points as sample data;
when the temperature point of the location is not the outer layer temperature point, performing step S7;
step S7, selecting variables including adjacent temperature points, internal temperature and humidity and moisture as sample data;
step S8, inputting the sample data into an input sequence of an input layer, and selecting the number of nodes of a hidden layer;
step S9, initializing weight and threshold, and setting training period and precision;
step S10, calculating the actual output of the network layer;
step S11, calculating an error, and correcting the weight and the threshold;
step S12, carrying out BP neural network training, and judging whether the training precision reaches a preset precision or whether the training period reaches a preset period;
when the training precision reaches the preset precision or/and the training period reaches the preset period, executing step S13;
step S13, predicting temperature values corresponding to all coordinates of the section in step S13;
when the training precision does not reach the preset precision and the training period does not reach the preset period, executing step S10;
and step S14, performing spatial interpolation according to the predicted point data to obtain the temperature field diagram of the cross section.
The invention reads the data in the database, intercepts different time point data, carries out 3D space reconstruction on the temperature, selects the section and predicts the plane temperature of the section, judges whether the temperature points at different positions in the section are outer layer temperature points or not, uses the selected variable as sample data to realize the preprocessing process of the granary data, rearranges the data to be used as a training set and a test set in a neural training network, then carries out neural network training to predict the plane temperature points, and finally interpolates the predicted plane temperature points, thereby predicting the real temperature field prediction diagram in the warehouse, solving the technical problems that the grain temperature prediction scheme of the existing granary can not predict the specific temperature distribution of the whole granary, and each granary is subjected to mathematical modeling analysis without being practical, and obtaining the space-time change trend of the temperature field in the future granary, and the temperature information in the granary stack can be intuitively and efficiently reflected, and the granary stack temperature information display system has universality and expandability and can be applied to various storage fields.
As a further improvement of the above scheme, the formula of the space-time prediction model is as follows:
in the formula, zt+1(ε) is the interpolated value with a spatial position of ε, zt+1(ai,bi,ci) Is a known position of (a)i,bi,ci) The interpolated value of (a)i,bi,ci) The grain bin is a line-row layer in the grain bin and corresponds to the actual position of the temperature sensor in the grain bin; theta 2]Representing a time-sequential prediction process from real data.
As a further improvement of the above scheme, before the training of the BP neural network, the data preprocessing is also performed on the temperature data: respectively corresponding different temperature points to each position of the granary, and carrying out data arrangement by taking field cable arrangement as a basis; the data sequence formula of the space temperature point is as follows:
X(a,b,c)＝[x1(a,b,c),x2(a,b,c),…,xt(a,b,c)]
wherein X (a, b, c) is a data sequence of temperature points of a layer in a row a, a column b and a column c in the granary, and XtAnd (a, b and c) are temperature point data of a layer c in a row a, a column b and a column in the granary, and t is the number of sampling data and also represents a time sequence.
As a further improvement of the above scheme, let l be the number of layers of the hidden layer of the BP neural network, n be the number of layers of the input layer of the BP neural network, and m be the number of layers of the output layer of the BP neural network. The output formula of the hidden layer of the BP neural network is as follows:
in the formula, xiIs an input value, omega, of an input layer of the BP neural networkijFor the weights of the input layer to the hidden layer, αjA bias for the input layer to the hidden layer; g (x) is an excitation function and satisfies:
as a further improvement of the above scheme, an output formula of an output layer of the BP neural network is:
in the formula (I), the compound is shown in the specification,
ξ is a constant number from 1 to 10jkWeight of the hidden layer to the output layer, βkIs the biasing of the hidden layer to the output layer.
Further, in step S11, the error calculation formula is:
in the formula, ykIs the desired output.
Still further, a modification formula of the weight from the hidden layer to the output layer is as follows:
the modification formula of the weight from the input layer to the hidden layer is as follows:
in the formula, ek＝yk-okAnd η is the learning rate.
Still further, the modification formula of the bias from the hidden layer to the output layer is:
the modification formula of the bias from the input layer to the hidden layer is as follows:
in the formula, ek＝yk-okAnd η is the learning rate.
As a further improvement of the above scheme, the spatial interpolation formula is:
wherein z (epsilon) is a numerical value at a point epsilon to be predicted in space, and z (a)i,bi,ci) The predicted value of the ith position of the actual temperature sensor in the granary is shown, w is the number of measured values, and lambda isiAs a weight coefficient, satisfy:
the invention also provides a prediction device of the granary space-time temperature field based on big data and interpolation prediction, which applies any of the prediction methods of the granary space-time temperature field based on big data and interpolation prediction, and comprises the following steps:
the data reading module is used for reading the grain condition data and corresponding warehouse information from a granary database;
the data interception module is used for intercepting different time point data according to the warehouse state in the warehouse information;
the reconstruction module is used for performing 3D space reconstruction on the temperature to obtain position information corresponding to the temperature point;
the section selection module is used for selecting a section in one direction and predicting the time sequence of the temperature of the section;
the temperature point judging module is used for judging whether the temperature points at different positions in the section are outer layer temperature points or not;
the first sample data selection module is used for selecting variables comprising meteorological data, internal temperature and humidity and adjacent temperature points as sample data when the temperature point of the position is the outer-layer temperature point;
the second sample data selection module is used for selecting variables including adjacent temperature points, internal temperature humidity and moisture as sample data when the temperature point of the position is not the outer-layer temperature point;
the sample data input module is used for inputting the sample data into an input sequence of an input layer and selecting the number of nodes of a hidden layer;
the initial setting module is used for initializing the weight and the threshold value and setting the training period and the precision;
a calculation module for calculating an actual output of the network layer;
the correcting module is used for calculating errors and correcting the weight and the threshold;
the training judgment module is used for carrying out BP neural network training and judging whether the training precision reaches a preset precision or not or judging whether the training period reaches a preset period or not; when the training precision does not reach the preset precision and the training period does not reach the preset period, the training judgment module drives the calculation module to calculate;
the prediction module is used for predicting temperature values corresponding to all coordinates of the cross section when the training precision reaches the preset precision or/and the training period reaches the preset period; and
and the temperature field acquisition module is used for carrying out spatial interpolation according to the predicted point data to obtain a temperature field diagram of the section.
Compared with the grain temperature prediction scheme of the existing granary, the granary space-time temperature field prediction method and device based on big data and interpolation prediction have the following beneficial effects:
the prediction method of the granary space-time temperature field based on big data and interpolation prediction comprises the steps of firstly reading data in a database, then intercepting different time point data, then carrying out 3D space reconstruction on the temperature, then selecting a section and predicting the plane temperature of the section, then judging whether temperature points at different positions in the section are outer temperature points or not, selecting a variable as sample data to realize the preprocessing process of the granary data, then rearranging the data to be used as a training set and a test set in a neural training network, then carrying out neural network training to predict the plane temperature points, and finally carrying out interpolation on the predicted plane temperature points, thereby predicting the real temperature field prediction graph in the granary, accurately predicting the variation trend of the temperature field in the granary, intuitively and effectively reflecting the internal temperature information of the granary pile, and realizing the universality and expandability of the model, can be applied to various storage sites.
Moreover, the prediction method of the empty-time temperature field of the granary adopts a BP neural network method to predict plane temperature points, further provides effective data for a temperature field prediction graph, adopts a Krigin interpolation method to perform spatial interpolation fitting to obtain a two-dimensional temperature field, establishes an empty-time temperature field prediction model, can intuitively and clearly reflect the internal temperature information of the granary, and provides a theoretical basis for sampling accuracy of suspicious points. Under the condition that the number of the sensors is limited, the method can reasonably analyze the temperature in the granary in space and time on the premise of reasonably arranging the temperature sensors, so that the visualization degree of the grain condition information is greatly improved, the reliability is increased while the improvement is realized, and the grain storage quality is guaranteed.
The beneficial effects of the prediction device of the granary space-time temperature field based on big data and interpolation prediction are the same as those of the prediction method of the granary space-time temperature field, and are not repeated herein.
Drawings
Fig. 1 is a flowchart of a prediction method of a granary space-time temperature field based on big data and interpolation prediction in embodiment 1 of the present invention.
Fig. 2 is a temperature prediction contrast diagram of coordinates (1,1,1) in an experiment of a prediction method of a granary space-time temperature field based on big data and interpolation prediction in embodiment 2 of the present invention.
Fig. 3 is a temperature prediction contrast diagram of coordinates (4,3,2) in an experiment of a prediction method of a granary space-time temperature field based on big data and interpolation prediction in embodiment 2 of the present invention.
Fig. 4 is a simulation comparison graph of the temperature field at the upper moment in the experiment and the predicted temperature field at the current moment in the granary space-time temperature field prediction method based on big data and interpolation prediction in embodiment 2 of the present invention.
Fig. 5 is a simulation comparison graph of the actual temperature field and the predicted temperature field at this moment in the experiment of the granary space-time temperature field prediction method based on big data and interpolation prediction in embodiment 2 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
Referring to fig. 1, the present embodiment provides a prediction method of a granary space-time temperature field based on big data and interpolation prediction, which combines big data analysis and spatial interpolation, and uses the data analysis method to realize omnibearing prediction of a temperature point in a granary in a time sequence, and then uses the spatial interpolation method to interpolate the temperature point in the granary, and uses a temperature field diagram to analyze the temperature in the granary. In this embodiment, the method for predicting the spatio-temporal temperature field of the granary adopts a BP neural network and a kriging interpolation method to establish a spatio-temporal temperature field model of the granary, and includes the following steps.
And step S1, the grain situation data and the corresponding warehouse information are read from a granary database. The granary database may be an existing database and store various information of the granary, such as grain condition data, warehouse information, management information, and the like.
And step S2, intercepting different time point data according to the warehouse state in the warehouse information. In this embodiment, the space-time temperature field model requires that the time-series prediction of the temperature points in the granary be performed first. Therefore, in this step, different time points are intercepted, so as to provide data support for subsequent processing.
And step S3, performing 3D space reconstruction on the temperature to obtain position information corresponding to the temperature point.
And step S4, selecting a section in one direction, and predicting the plane temperature of the section. In this embodiment, the temperature field to be displayed finally is a section of the granary selected as a temperature display surface, so as to predict more comprehensive and abundant temperature data.
And step S5, judging whether the temperature points at different positions in the cross section are outer layer temperature points.
When the temperature point of the location is the outer layer temperature point, step S6 is performed.
In step S6, variables including meteorological data, internal temperature and humidity, and adjacent temperature points are selected as sample data.
When the temperature point of the position is not the outer layer temperature point, step S7 is performed.
In step S7, variables including the proximity temperature point, the internal temperature and humidity, and the moisture are selected as sample data.
In this embodiment, the timing prediction method is based on a BP neural network, which is composed of an input layer, a hidden layer, and an output layer. And inputting an input sequence of the layers, namely temperature data and humidity data of the granary. Data taken out of the database needs to be preprocessed and then used as an input sequence of the neural network. In this embodiment, the temperature data also needs to be preprocessed: respectively corresponding different temperature points to each position of the granary, and carrying out data arrangement by taking field cable arrangement as a basis; the data sequence formula of the space temperature point is as follows:
X(a,b,c)＝[x1(a,b,c),x2(a,b,c),…,xt(a,b,c)]
wherein X (a, b, c) is a data sequence of temperature points of a layer in a row a, a column b and a column c in the granary, and XtAnd (a, b and c) are temperature point data of a layer c in a row a, a column b and t is the number of sampling data. Since the data taken is timing data, t is also time-sequential.
And after the temperature data is processed, training the BP neural network by combining the humidity data. The training process consists of two stages, forward propagation of the signal and back propagation of the error. The former stage starts from the input layer, the output layer ends, and the middle passes through the hidden layer. In the latter stage, the output layer is started, the input layer is terminated, the middle layer passes through the hidden layer, but the weight and the bias in the propagation process need to be adjusted respectively.
Step S8, inputting the sample data as an input sequence of the input layer, and selecting the number of hidden layer nodes.
In this embodiment, the signal propagates forward, and the output formula of the hidden layer of the BP neural network is:
in the formula, xiIs BP neural netInput value of the input layer of the complex, omegaijFor input layer to hidden layer weights, αjIs the biasing of the input layer to the hidden layer. g (x) is an excitation function and satisfies:
in this embodiment, the output formula of the output layer of the BP neural network is:
in the formula, l is the number of hidden layers, and satisfies the following conditions:
n is the number of input layers, m is the number of output layers, ξ is a constant of 1-10jkFor hidden layer to output layer weights, βkIs the bias of the hidden layer to the output layer.
And step S9, initializing the weight and the threshold, and setting the training period and the precision.
Step S10, the actual output of the network layer is calculated.
And step S11, calculating errors, and correcting the weight and the threshold.
In this embodiment, the error propagates in the opposite direction, so the error calculation formula is:
in the formula, ykIs the desired output. Wherein i is 1,2, …, n; j ═ 1,2, … l; k is 1,2, … m.
In this embodiment, the weight is adjusted by using a gradient descent method, and a modification formula of the weight from the hidden layer to the output layer is as follows:
the modification formula of the weight from the input layer to the hidden layer is as follows:
in the formula, ek＝yk-okAnd η is the learning rate.
And, the modification formula for the bias from the hidden layer to the output layer is:
the correction formula for the bias of the input layer to the hidden layer is:
in the formula, ek＝yk-okAnd η is the learning rate.
Step S12, performing BP neural network training, and determining whether the training precision reaches a preset precision, or determining whether the training period reaches a preset period.
When the training precision reaches the preset precision or/and the training period reaches the preset period, step S13 is executed.
And step S13, predicting temperature values corresponding to all coordinates of the cross section.
When the training accuracy does not reach the preset accuracy and the training period does not reach the preset period, step S10 is executed.
And after the predicted value of the temperature point of the granary is obtained, carrying out the next step, and carrying out spatial interpolation on the discrete temperature point of the granary so as to obtain a temperature predicted value around the temperature measuring point of the granary.
And step S14, performing spatial interpolation according to the predicted point data to obtain a temperature field diagram of the cross section.
In this embodiment, the spatiotemporal analysis of the temperature field is used in the storage of the granary, the granary is spatially reconstructed, the data of each temperature point is corresponding to each position of the granary, the cable arrangement on the site is used as a standard, the row and column layers of the granary are represented by a, b and c, the temperature data is predicted in time sequence, and then the spatial interpolation estimation is performed. The formula of the space-time prediction model is as follows:
in the formula, zt+1(ε) is the interpolated value with a spatial position of ε, zt+1(ai,bi,ci) Is a known position of (a)i,bi,ci) The interpolated value of (a)i,bi,ci) The grain bin is a line-row layer in the grain bin and corresponds to the actual position of the temperature sensor in the grain bin; theta 2]Representing a time series prediction process based on real data, associated with the prior granary temperature data.
The temperature field to be displayed finally in the model is that a certain section of the granary is selected as a temperature display surface, points on the current section are selected for space-time prediction analysis, and then fitting is carried out to obtain a two-dimensional graph of the temperature field. In this embodiment, for the trained BP neural network, the input temperature data is predicted, and the predicted value output by the neural network is
Representing the prediction data of each temperature point of the granary. The discrete temperature points of the planar distribution are then interpolated. In this embodiment, the temperature points in the plane are equivalent to the observation points, the surrounding blank is the non-sampling points, the values of the non-sampling points are the linear weighted average of the adjacent observation values, and the weight is determined by the fitted variation function. Thus, the spatial interpolation formula is:
where z (ε) is the value at the point ε in space that needs to be predicted, and is obtained by linear combination of w observed sample values. Epsilon here is any position in the granary space and is not limited to the temperature points (a, b, c). z (a)i,bi,ci) For the actual temperature sensor in the granaryPredicted value at the ith position, w is the number of measurements, λiAs a weight coefficient, satisfy:
λithe value of (A) directly determines the precision of the value to be estimated, and because the basis of the kriging interpolation method is the unbiased optimal estimation, the lambda is determinediSatisfy the requirement of
Z(s) is satisfied if the estimated value is only related to the known values of the adjacent positions0) Covariance of (c) Cov(s)i,si) Exists of λiThe variance of the increment is required to be minimum, so that the correlation between the estimated data and the known data is strongest, and the variance is minimum.
In summary, compared with the grain temperature prediction scheme of the existing granary, the granary space-time temperature field prediction method based on big data and interpolation prediction of the embodiment has the following advantages:
the prediction method of the granary space-time temperature field based on big data and interpolation prediction comprises the steps of firstly reading data in a database, then intercepting different time point data, then carrying out 3D space reconstruction on the temperature, then selecting a section and predicting the plane temperature of the section, then judging whether temperature points at different positions in the section are outer temperature points or not, selecting a variable as sample data to realize the preprocessing process of the granary data, then rearranging the data to be used as a training set and a test set in a neural training network, then carrying out neural network training to predict plane temperature points, and finally carrying out interpolation on the predicted plane temperature points, thereby predicting the real temperature field prediction graph in the granary, accurately predicting the variation trend of the temperature field in the granary, intuitively and effectively reflecting the internal temperature information of the granary pile, and realizing the universality and expandability of a model, can be applied to various storage sites.
Moreover, the prediction method of the empty-time temperature field of the granary adopts a BP neural network method to predict plane temperature points, further provides effective data for a temperature field prediction graph, adopts a Krigin interpolation method to perform spatial interpolation fitting to obtain a two-dimensional temperature field, establishes an empty-time temperature field prediction model, can intuitively and clearly reflect the internal temperature information of the granary, and provides a theoretical basis for sampling accuracy of suspicious points. Under the condition that the number of the sensors is limited, the method can reasonably analyze the temperature in the granary in space and time on the premise of reasonably arranging the temperature sensors, so that the visualization degree of the grain condition information is greatly improved, the reliability is increased while the improvement is realized, and the grain storage quality is guaranteed.
Example 2
The embodiment provides a prediction method of a granary space-time temperature field based on big data and interpolation prediction, and a simulation experiment is performed on the basis of the embodiment 1. And taking a bin under a certain grain condition user from a grain condition database for temperature field prediction, wherein the data range is historical data of the last half year, and the data range is 328 data in total. The maximum training times of the BP neural network are set to be 5000 times, the learning rate is 0.05, and the required training precision is 1 e-3. The row-column layer of the warehouse is 7 rows, 6 columns and 4 layers.
First, the historical data of the warehouse is taken out for processing. And for the temperature field section to be predicted, acquiring corresponding variable data as an input sequence of the BP neural network according to the position of the temperature point. Such as: for the prediction of the outer-layer temperature point with the coordinate point of (1,1,1), due to the fact that the outer-layer temperature point is close to the wall or the top of the granary, the influence factors comprise outer temperature and outer humidity, inner temperature and inner humidity, and temperature data of adjacent points (1,1,2) and (1,2, 1); for the temperature point at the center, such as the temperature prediction of the coordinate point (4,3,2), the influence of external temperature and external humidity is small, the correlation between the internal temperature and internal humidity and the temperature points at the front, back, left, right, upper and lower parts is large, and the internal temperature and internal humidity data and the temperature data of the adjacent points (4,3,3), (4,3,1), (4,2,2), (4,4,2), (5,3,2) and (3,3,2) can be acquired.
And analyzing and processing the data by adopting a BP neural network method, setting a training period and precision, training a training set, testing by using a test set, and drawing and displaying points at different positions in the granary, wherein the results are shown in fig. 2 and fig. 3.
Selecting historical data of a corresponding plane, predicting through a BP neural network, obtaining predicted values of temperature points at different positions in the plane, and comparing the predicted values with the actual values. The section of the first row, i.e. the comparison table of the predicted values of the actual values of the temperature points with coordinates (1, b, c), is shown in the following table, please refer to table 1.
TABLE 1 prediction table of section temperature values
And finally, after a temperature predicted value of the current plane is obtained, restoring the spatial distribution of the temperature points to an actual scene by combining actual warehouse information, and performing Krigin interpolation fitting according to the predicted value of the temperature points to obtain the temperature field distribution of the plane. According to the warehouse arrangement, the length of the warehouse is 40 meters, and the width of the warehouse is 35 meters. The distance between the rows and the columns is 5 meters, and the distance between the outermost layer of cables and the surrounding walls is 2.5 meters; the distance between the layers is 1.5 m, and the distance between the topmost temperature measuring point and the top layer is 0.75 m. The temperature measuring points are arranged from top to bottom, coordinates (1,1) on the cross section correspond to (0.75,5.25) in the cross section of the actual warehouse, and the drawing is performed by analogy, as shown in fig. 4 and 5.
Furthermore, the model is evaluated using the root mean square error and the mean absolute percentage error. For the experimental data, the predicted temperature value and the actual value corresponding to the position of the measurement and control point in the granary are subjected to error detection to obtain: RMSE 0.1387; MAPE ═ 1.75%. The results show that the temperature field prediction graph obtained by the method adopted in the example has good effect.
Example 3
The embodiment provides a granary space-time temperature field prediction device based on big data and interpolation prediction, which applies the granary space-time temperature field prediction method based on big data and interpolation prediction of embodiment 1, and comprises a data reading module, a data intercepting module, a reconstruction module, a cross section selecting module, a temperature point judging module, a sample data selecting module I, a sample data selecting module II, a sample data input module, an initial setting module, a calculating module, a correcting module, a training judging module, a prediction module and a temperature field obtaining module.
The data reading module is configured to read the grain situation data and the corresponding warehouse information from a grain warehouse database, and may implement step S1 in embodiment 1. The data intercepting module is configured to intercept different point-in-time data according to the warehouse status in the warehouse information, and the module may implement step S2 in embodiment 1. The reconstruction module is configured to perform 3D spatial reconstruction on the temperature to obtain position information corresponding to the temperature point, and the module may implement step S3 in embodiment 1. The section selection module is used to select an azimuth section and predict the plane temperature of the section, and the module can implement step S4 in embodiment 1. The temperature point determination module is configured to determine whether the temperature points at different positions in the cross section are outer layer temperature points, and the module may implement step S5 in embodiment 1. The first sample data selection module is configured to select variables including meteorological data, internal temperature and humidity, and adjacent temperature points as sample data when the real-time temperature point is the outer temperature point, and the module may implement step S6 in embodiment 1. The second sample data selection module is configured to select variables including the neighboring temperature point, the internal temperature, the internal humidity, and the moisture as sample data when the real-time temperature point is not the outer-layer temperature point, and the module may implement step S7 in embodiment 1. The sample data input module is configured to input sample data as an input sequence of an input layer, and select a number of nodes of an implicit layer, and the module may implement step S8 in embodiment 1. The initial setting module is used to initialize the weight and the threshold, and set the training period and the precision, and this module may implement step S9 in embodiment 1. The calculation module is used to calculate the actual output of the network layer, and the module may implement step S10 in embodiment 1. The modification module is used for calculating an error and modifying the weight and the threshold, and the module may implement step S11 in embodiment 1. The training determination module is configured to perform BP neural network training, and determine whether the training precision reaches a preset precision, or determine whether the training period reaches a preset period, where the training determination module may implement step S12 in embodiment 1. And when the training precision does not reach the preset precision and the training period does not reach the preset period, the training judgment module drives the calculation module to calculate. The prediction module is configured to predict temperature values corresponding to all coordinates of the cross section when the training precision reaches a preset precision or/and the training period reaches a preset period, and the module may implement step S13 in embodiment 1. The temperature field obtaining module is configured to perform spatial interpolation according to the predicted point data to obtain a temperature field map of the cross section, and the module may implement step S14 in embodiment 1.
The advantage of the prediction apparatus of the granary space-time temperature field based on big data and interpolation prediction in this embodiment is the same as the beneficial effect of the prediction method of the granary space-time temperature field, and is not described herein again.
Example 4
The present embodiments provide a computer terminal comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. The steps of the prediction method of the granary space-time temperature field based on big data and interpolation prediction in embodiment 1 are realized when the processor executes the program.
In the embodiment 1, when the prediction method of the empty-time temperature field of the granary is applied, the prediction method can be applied in a software mode, for example, a program designed to run independently is installed on a computer terminal, and the computer terminal can be a computer, a smart phone, a control system, other internet-of-things equipment and the like. The prediction method of the empty-time temperature field of the granary in the embodiment 1 can also be designed into an embedded running program and installed on a computer terminal, such as a single chip microcomputer.
Example 5
The present embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the prediction method of the granary space-time temperature field based on big data and interpolation prediction of the embodiment 1 are realized.
When the method for predicting the empty-time temperature field of the grain bin in embodiment 1 is applied, the method can be applied in the form of software, for example, a program which is designed to be independently run by a computer-readable storage medium, the computer-readable storage medium can be a usb disk which is designed as a usb shield, and the usb disk is designed to be a program which starts the whole method by external triggering.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A granary space-time temperature field prediction method based on big data and interpolation prediction is characterized by comprising the following steps:
step S1, reading grain situation data and corresponding warehouse information from a granary database;
step S2, intercepting different time point data according to the warehouse state in the warehouse information;
step S3, performing 3D space reconstruction on the temperature to obtain position information corresponding to the temperature point;
step S4, selecting a section to be predicted, and firstly performing time sequence prediction of the section temperature;
step S5, judging whether the temperature points of different positions of the cross section are outer layer temperature points or not;
when the temperature point of the location is the outer layer temperature point, performing step S6;
step S6, selecting variables including meteorological data, internal temperature and humidity and adjacent temperature points as sample data;
when the temperature point of the location is not the outer layer temperature point, performing step S7;
step S7, selecting variables including adjacent temperature points, internal temperature and humidity and moisture as sample data;
step S8, inputting the sample data into an input sequence of an input layer, and selecting the number of nodes of a hidden layer;
step S9, initializing weight and threshold, and setting training period and precision;
step S10, calculating the actual output of the network layer;
step S11, calculating an error, and correcting the weight and the threshold;
step S12, carrying out BP neural network training, and judging whether the training precision reaches a preset precision or whether the training period reaches a preset period;
when the training precision reaches the preset precision or/and the training period reaches the preset period, executing step S13;
step S13, predicting temperature values corresponding to all coordinates of the section in step S13;
when the training precision does not reach the preset precision and the training period does not reach the preset period, executing step S10;
and step S14, performing spatial interpolation according to the predicted point data to obtain the temperature field diagram of the cross section.
2. The method of predicting a spatio-temporal temperature field of a granary based on big data and interpolation prediction according to claim 1, wherein the formula of the spatio-temporal prediction model is:
in the formula, zt+1(epsilon) is an interpolation value with a spatial position of epsilon; z is a radical oft+1(ai,bi,ci) Is a known position of (a)i,bi,ci) The interpolated value of (d); (a)i,bi,ci) The grain bin is a line-row layer in the grain bin and corresponds to the actual position of the temperature sensor in the grain bin; theta 2]Representing a time-sequential prediction process from real data.
3. The method for predicting the spatiotemporal temperature field of the granary based on the big data and the interpolation prediction as claimed in claim 1, wherein before the training of the BP neural network, the data preprocessing is further performed on the temperature data: respectively corresponding different temperature points to each position of the granary, and carrying out data arrangement by taking field cable arrangement as a basis; the data sequence formula of the space temperature point is as follows:
X(a,b,c)＝[x1(a,b,c),x2(a,b,c),…,xt(a,b,c)]
wherein X (a, b, c) is a data sequence of temperature points of a layer in a row a, a column b and a column c in the granary, and Xt(a, b, c) is the temperature point data of the layer c in the row a, column b and column a in the granary, and t is the number of the sampling dataNumbers, also represent time series.
4. The method according to claim 1, wherein l is the number of layers of the hidden layer of the BP neural network, n is the number of layers of the input layer of the BP neural network, and m is the number of layers of the output layer of the BP neural network. The method is characterized in that the output formula of the hidden layer of the BP neural network is as follows:
in the formula, xiThe input value of the input layer of the BP neural network is x at different positionst(a,b,c)；ωijα as the weight of the input layer to the hidden layerjA bias for the input layer to the hidden layer; g (x) is an excitation function and satisfies:
5. the method of predicting the spatiotemporal temperature field of the granary based on big data and interpolation prediction according to claim 4, wherein the output formula of the output layer of the BP neural network is as follows:
in the formula, satisfy:
ξ is a constant number from 1 to 10jkWeight of the hidden layer to the output layer, βkIs the biasing of the hidden layer to the output layer.
6. The method of predicting the spatiotemporal temperature field of a granary based on big data and interpolation prediction according to claim 5, wherein in step S12, the error calculation formula is:
in the formula, ykIs the desired output.
7. The method of claim 6, wherein the formula for modifying the weights from the hidden layer to the output layer is as follows:
the modification formula of the weight from the input layer to the hidden layer is as follows:
in the formula, ek＝yk-okAnd η is the learning rate.
8. The method of claim 6, wherein the offset from the hidden layer to the output layer is modified by the formula:
the modification formula of the bias from the input layer to the hidden layer is as follows:
in the formula, ek＝yk-okAnd η is the learning rate.
9. The method of claim 1, wherein the spatial interpolation formula is as follows:
wherein z (epsilon) is a numerical value at a point epsilon to be predicted in space, and z (a)i,bi,ci) The predicted value of the ith position of the actual temperature sensor in the granary is shown, w is the number of measured values, and lambda isiAs a weight coefficient, satisfy:
10. a prediction device of a granary space-time temperature field based on big data and interpolation prediction, which is applied to the prediction method of the granary space-time temperature field based on big data and interpolation prediction according to any one of claims 1-9, and is characterized by comprising:
the data reading module is used for reading the grain condition data and corresponding warehouse information from a granary database;
the data interception module is used for intercepting different time point data according to the warehouse state in the warehouse information;
the reconstruction module is used for performing 3D space reconstruction on the temperature to obtain position information corresponding to the temperature point;
the section selection module is used for selecting a section in one direction and predicting the time sequence of the temperature of the section;
the temperature point judging module is used for judging whether the temperature points at different positions in the section are outer layer temperature points or not;
the first sample data selection module is used for selecting variables comprising meteorological data, internal temperature and internal humidity and adjacent temperature points as sample data when the real-time temperature point is the outer-layer temperature point;
the second sample data selection module is used for selecting variables including adjacent temperature points, internal temperature humidity and moisture as sample data when the real-time temperature point is not the outer-layer temperature point;
the sample data input module is used for inputting the sample data into an input sequence of an input layer and selecting the number of nodes of a hidden layer;
the initial setting module is used for initializing the weight and the threshold value and setting the training period and the precision;
a calculation module for calculating an actual output of the network layer;
the correcting module is used for calculating errors and correcting the weight and the threshold;
the training judgment module is used for carrying out BP neural network training and judging whether the training precision reaches a preset precision or not or judging whether the training period reaches a preset period or not; when the training precision does not reach the preset precision and the training period does not reach the preset period, the training judgment module drives the calculation module to calculate;
the prediction module is used for predicting temperature values corresponding to all coordinates of the cross section when the training precision reaches the preset precision or/and the training period reaches the preset period; and
and the temperature field acquisition module is used for carrying out spatial interpolation according to the predicted point data to obtain a temperature field diagram of the section.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010079018.0A CN111310317B (en) | 2020-02-03 | 2020-02-03 | Granary space-time temperature field prediction method and device based on big data and interpolation prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010079018.0A CN111310317B (en) | 2020-02-03 | 2020-02-03 | Granary space-time temperature field prediction method and device based on big data and interpolation prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111310317A true CN111310317A (en) | 2020-06-19 |
| CN111310317B CN111310317B (en) | 2023-02-14 |
Family
ID=71147013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010079018.0A Active CN111310317B (en) | 2020-02-03 | 2020-02-03 | Granary space-time temperature field prediction method and device based on big data and interpolation prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111310317B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112577671A (en) * | 2020-11-27 | 2021-03-30 | 武汉工程大学 | Well lid monitoring method and system by using Kriging method |
| CN113092178A (en) * | 2021-06-09 | 2021-07-09 | 亿海蓝（北京）数据技术股份公司 | Control method and control system of sampling machine and readable storage medium |
| CN116227247A (en) * | 2023-05-09 | 2023-06-06 | 交通运输部水运科学研究所 | Grain silo reverse simulation method and system based on big data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015172560A1 (en) * | 2014-05-16 | 2015-11-19 | 华南理工大学 | Central air conditioner cooling load prediction method based on bp neural network |
| CN106227038A (en) * | 2016-07-29 | 2016-12-14 | 中国人民解放军信息工程大学 | Grain drying tower intelligent control method based on neutral net and fuzzy control |
| CN109639762A (en) * | 2018-11-07 | 2019-04-16 | 重庆光电信息研究院有限公司 | City Internet of Things information grading processing system and method |
-
2020
- 2020-02-03 CN CN202010079018.0A patent/CN111310317B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015172560A1 (en) * | 2014-05-16 | 2015-11-19 | 华南理工大学 | Central air conditioner cooling load prediction method based on bp neural network |
| CN106227038A (en) * | 2016-07-29 | 2016-12-14 | 中国人民解放军信息工程大学 | Grain drying tower intelligent control method based on neutral net and fuzzy control |
| CN109639762A (en) * | 2018-11-07 | 2019-04-16 | 重庆光电信息研究院有限公司 | City Internet of Things information grading processing system and method |
Non-Patent Citations (1)
| Title |
|---|
| 高松等: "基于BP神经网络方法的高大平房仓温度场预测研究", 《粮油食品科技》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112577671A (en) * | 2020-11-27 | 2021-03-30 | 武汉工程大学 | Well lid monitoring method and system by using Kriging method |
| CN113092178A (en) * | 2021-06-09 | 2021-07-09 | 亿海蓝（北京）数据技术股份公司 | Control method and control system of sampling machine and readable storage medium |
| CN116227247A (en) * | 2023-05-09 | 2023-06-06 | 交通运输部水运科学研究所 | Grain silo reverse simulation method and system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111310317B (en) | 2023-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111310317B (en) | Granary space-time temperature field prediction method and device based on big data and interpolation prediction | |
| Yañez‐Arenas et al. | Modelling geographic patterns of population density of the white‐tailed deer in central Mexico by implementing ecological niche theory | |
| US5140523A (en) | Neural network for predicting lightning | |
| Danciu et al. | The 2014 Earthquake Model of the Middle East: ground motion model and uncertainties | |
| Chan et al. | Detecting concerted demographic response across community assemblages using hierarchical approximate Bayesian computation | |
| JPWO2016031174A1 (en) | Simulation device, simulation method, and storage medium | |
| CN110533173B (en) | Multi-level attention network-based storage grain pile temperature prediction method and device | |
| Stein et al. | Metrics for assessing earthquake‐hazard map performance | |
| CN113901384A (en) | Ground PM2.5 concentration modeling method considering global spatial autocorrelation and local heterogeneity | |
| Pulido Fentanes et al. | Kriging‐based robotic exploration for soil moisture mapping using a cosmic‐ray sensor | |
| CN115935139A (en) | Space field interpolation method for ocean observation data | |
| CN110161181A (en) | The concentration of component recognition methods of mixed gas and system | |
| CN110490061B (en) | Uncertainty modeling and measuring method for remote sensing image characteristics | |
| CN110188399B (en) | Dam safety monitoring single-measuring-point evaluation method based on multiple correlation sequences | |
| CN115640860A (en) | Electromechanical equipment remote maintenance method and system for industrial cloud service | |
| CN104768175B (en) | A kind of Local Area Network coverage detection method and system | |
| KR101635791B1 (en) | Determination method for location and origin time of earthquake using arrival time of primary wave | |
| JP2021154935A (en) | Vehicle simulation system, vehicle simulation method and computer program | |
| Ortega et al. | Automatic selection of dispersion curves based on a weighted probability scheme | |
| Amboni et al. | The effect of species geographical distribution estimation methods on richness and phylogenetic diversity estimates | |
| US11741113B2 (en) | Measurement guide device and simulation computing device used therefor | |
| CN112528417B (en) | Assessment method for semi-physical simulation of aircraft | |
| CN112819813B (en) | Intelligent underground pipeline identification method and device and storage medium | |
| CN113569324A (en) | Slope deformation monitoring abnormal data analysis and optimization method | |
| Gilleland | Package ‘verification’ |
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 |