Multi-information fusion large-scale space temperature and humidity evaluation method

Abstract

The invention discloses a multi-information fusion large-scale space temperature and humidity evaluation method, and the method comprises the steps: uniformly dividing a large-scale space into a plurality of regional spaces, and arranging a standard sensor in each regional space; setting the same sampling frequency, recording the data of the detected temperature and humidity sensor, collecting the data of the standard sensor to obtain a related data sequence, and calculating the relevance between the data of the detected temperature and humidity sensor and the data of the standard sensor; according to the calculation result of the relevance, data of the relevant detected temperature and humidity sensor is reserved, and an ARIMA model based on a time sequence is established; calculating a weighted average coefficient according to a correlation result, and establishing a standard mathematical model in combination with the ARIMA model; therefore, the error condition of the detected temperature and humidity sensor is deduced. According to the method, the temperature and humidity change of each region in the next stage can be predicted, and error comparison and evaluation work can be carried out on a large-space temperature and humidity sensor.

Description

Technical Field

[0001] The invention relates to the technical field of metrology and calibration, and in particular to a large-scale spatial temperature and humidity assessment method fused with multiple information. Background Art

[0002] At present, the temperature and humidity detection in large-scale spaces is a complex working condition, which is mainly manifested in the following aspects: (1) The temperature and humidity sensors cannot be disassembled. For example, in a production workshop used to monitor temperature and humidity, or in a grain warehouse used to store grain, there is generally only one temperature and humidity sensor, which cannot be disassembled; (2) There are no corresponding technical specifications to refer to. For example, a grain warehouse has a large space size, but there are no corresponding technical specifications to refer to when detecting the temperature and humidity of its actual environment space. Summary of the invention

[0003] In order to solve the above problems, the present invention proposes a large-scale spatial temperature and humidity assessment method based on multi-information fusion, which can perform accurate temperature and humidity assessment in large-scale spaces and under conditions where the sensors cannot be removed.

[0004] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0005] The present invention is a large-scale spatial temperature and humidity assessment method based on multi-information fusion, comprising:

[0006] The entire space size is evenly divided according to the length and width of the plane size to obtain multiple regional spaces, and the standard sensors are arranged in each regional space;

[0007] Set the same sampling frequency, record the data of the tested temperature and humidity sensor, collect the data of the standard sensor, obtain the relevant data sequence, and calculate the correlation between the data of the tested temperature and humidity sensor and the data of the standard sensor;

[0008] According to the calculation result of the correlation, the data of the relevant temperature and humidity sensors under inspection are retained, and the ARIMA model of the time series is established according to the retained data of the temperature and humidity sensors under inspection;

[0009] Calculating the weighted average coefficient of correlation of the retained data of the inspected temperature and humidity sensors;

[0010] According to the ARIMA model of the time series and the calculated weighted average coefficient, a standard mathematical model is established;

[0011] The error of the tested temperature and humidity sensor is inferred based on the standard mathematical model.

[0012] A further improvement of the present invention is that the number of regional spaces is not less than 9.

[0013] A further improvement of the present invention is that the calculation expression of the correlation between the data of the tested temperature and humidity sensor and the data of the standard sensor is:

[0014] ;

[0015] in, is the comprehensive correlation, is the relative correlation, is the absolute correlation, is the correlation coefficient.

[0016] A further improvement of the present invention is that the calculation process of the absolute correlation degree is:

[0017] Compute the initial image of each data sequence:

[0018] ;

[0019] ;

[0020] in, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, is the data sequence of the temperature and humidity sensor under test, is the standard sensor data sequence, It is the first data of the data sequence of the temperature and humidity sensor under test. The first data of the standard sensor data sequence is is the initial image of the temperature and humidity sensor under test data, is the initial image of the standard sensor data, is the number of the tested temperature and humidity sensor data sequence or the standard sensor data sequence;

[0021] Calculate the distance between data series and :

[0022] ;

[0023] ;

[0024] in, is the spacing of the initial image data sequence of the temperature and humidity sensor under inspection, is the spacing of the initial image data sequence of the standard sensor, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, not subscript The serial number of the data sequence of the temperature and humidity sensor under test or the data sequence of the standard sensor;

[0025] Calculate the relative correlation:

[0026] .

[0027] A further improvement of the present invention is that the standard for retaining the data of the relevant temperature and humidity sensors being tested is high correlation, which is expressed as:

[0028] ;

[0029] in, The comprehensive correlation.

[0030] A further improvement of the present invention is that the expression of the ARIMA model of the time series is:

[0031] ;

[0032] in, is the predicted value, is the autoregressive term coefficient, is the moving average term coefficient, p is the autoregressive term, q is the moving average term, , is the previous value, is the interference term, For time series.

[0033] A further improvement of the present invention is that the calculation expression of the weighted average coefficient of the correlation of the data of the detected temperature and humidity sensors retained is:

[0034] ;

[0035] in, is the weighted average coefficient, is the comprehensive correlation of the data of the detected temperature and humidity sensors that are retained, is the number of standard sensors that are highly correlated with the temperature and humidity sensor data being tested, It is the serial number of the data of the detected temperature and humidity sensor that is retained.

[0036] A further improvement of the present invention is that the expression of the standard mathematical model is:

[0037] ;

[0038] in, is the weighted average coefficient, is the predicted value, For the standard model.

[0039] The beneficial effects of the present invention are as follows: the present invention can predict the temperature and humidity changes in each area in the next stage. In the case of a large-sized space and non-detachable sensors, the multi-information fusion model can be used as a standard model to carry out error comparison and evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Schematic diagram of the method structure framework of an embodiment of the present invention;

[0041] Figure 2 is a model diagram of the collected data in the embodiment of the present invention;

[0042] Figure 3 is a waveform diagram of humidity collected by the standard sensor 1 in the embodiment of the present invention;

[0043] Figure 4 This is a diagram of the difference result of the standard sensor 1 in the embodiment of the present invention;

[0044] Figure 5 is the ARIMA (7, 1, 7) prediction graph of the standard sensor 1 in the embodiment of the present invention;

[0045] Figure 6 It is a standard mathematical model and a waveform diagram of the instrument under test in the embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0047] like Figure 1 As shown, the large-scale spatial temperature and humidity assessment method of multi-information fusion in this embodiment includes:

[0048] Step 1: In the large-scale space, the entire space size is evenly divided according to the length and width of the plane size to obtain n regional spaces, and the standard sensors are arranged in each regional space so that the number of standard sensors in the entire large space is not less than 9.

[0049] Step 2: Set the same sampling frequency, record the data of the tested temperature and humidity sensor, collect the data of the standard sensor, record and collect once every set time, obtain the relevant data sequence, calculate the correlation between the data of the tested temperature and humidity sensor and the data of the standard sensor, and the calculation expression is:

[0050] (1);

[0051] in, is the comprehensive correlation, is the relative correlation, is the absolute correlation, is the correlation coefficient, which is generally taken as 0.5.

[0052] The calculation process of relative correlation is:

[0053] Calculate the initial image of each data sequence, that is, the difference processing of the data:

[0054] (2);

[0055] (3);

[0056] in, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, is the data sequence of the temperature and humidity sensor under test, is the standard sensor data sequence, It is the first data of the data sequence of the temperature and humidity sensor under test. The first data of the standard sensor data sequence is is the initial image of the temperature and humidity sensor under test data, is the initial image of the standard sensor data, is the number of the tested temperature and humidity sensor data sequence or the standard sensor data sequence;

[0057] Calculate the distance between data series and :

[0058] (4);

[0059] (5);

[0060] in, is the spacing of the initial image data sequence of the temperature and humidity sensor under inspection, is the spacing of the initial image data sequence of the standard sensor, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, not subscript The serial number of the data sequence of the temperature and humidity sensor under test or the data sequence of the standard sensor;

[0061] Calculate the relative correlation:

[0062] (6).

[0063] Calculate absolute correlation ,Right now:

[0064] (7);

[0065] (8);

[0066] in, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, is the data sequence of the detected sensor, It is the first data of the data sequence of the temperature and humidity sensor under test. is the initial image of the temperature and humidity sensor under test data, is the standard sensor data sequence, It is the first data in the standard sensor data sequence. is the initial image of the standard sensor data.

[0067] Calculate the distance between data series and :

[0068] (9);

[0069] (10);

[0070] in, is the initial image data sequence spacing of the temperature and humidity sensor under inspection, is the spacing of the initial image data sequence of the standard sensor, is the initial image of the temperature and humidity sensor under inspection, This is the initial image of the standard sensor.

[0071] The calculation expression of absolute correlation is:

[0072] (11).

[0073] Step 3: According to the calculation result of the correlation, the data of the relevant temperature and humidity sensors to be inspected are retained, and the ARIMA model of the time series is established based on the retained data of the temperature and humidity sensors to be inspected. The correlation between the data is judged by the following evaluation criteria:

[0074] (12).

[0075] The data of the highly correlated temperature and humidity sensors under inspection are retained, and the ARIMA model of the time series, namely the ARIMA (p, d, q) model, is established based on the retained data of the temperature and humidity sensors under inspection:

[0076] (13);

[0077] in, is the predicted value, is the autoregressive term coefficient, is the moving average term coefficient, p is the autoregressive term, q is the moving average term, , is the previous value, is the interference term, For time series.

[0078] Step 4, calculate the weighted average coefficient of the correlation of the data of the retained temperature and humidity sensors under inspection, and the calculation expression is:

[0079] (14);

[0080] in, is the weighted average coefficient, is the correlation coefficient of the retained data of the tested temperature and humidity sensors, is the number of standard sensors that are highly correlated with the temperature and humidity sensor data being tested, It is the serial number of the data of the detected temperature and humidity sensor that is retained.

[0081] Step 5: Based on the ARIMA of the time series and the calculated weighted average coefficient, a standard mathematical model is established, expressed as:

[0082] (15);

[0083] in, is the weighted average coefficient, is the predicted value, For the standard model.

[0084] Step 6: Predict the humidity change of each regional space based on the established ARIMA of the time series of each regional space, and infer the error of the tested temperature and humidity sensor based on the standard mathematical model.

[0085] This embodiment detects the humidity of a granary. According to the size of the granary, 13 areas are divided and 13 standard sensors are placed. Data is collected every 1 minute to obtain humidity data such as Figure 2 shown.

[0086] The grey comprehensive correlation between the humidity data collected by the standard sensor and the data of the sensor under test is calculated, and the results are shown in Table 1 below:

[0087] Table 1 Grey correlation results

[0088]

[0089] According to the principle that the comprehensive correlation is greater than or equal to 0.85, the data of standard sensors 3, 7, 8, 9, 12, and 13 are eliminated, and the valid data of standard sensors 1, 4, 5, 6, 10, and 11 are retained.

[0090] A time series ARIMA model is established for a total of 5 groups of data of the retained standard sensors 1, 4, 5, 6, 10, and 11. Taking standard sensor 1 as an example, the process of establishing the ARIMA (p, d, q) model is as follows:

[0091] 1. First, check the stability of the humidity data collected by standard sensor 1. Figure 3 shown.

[0092] 2. The above data fluctuates greatly, so it is necessary to perform differential processing on the above data. After one differential, the data becomes stable, and the results are as follows: Figure 4 As shown, we can also get d=1.

[0093] 3. Establish an ARIMA (p, d, q) model for the differenced data. After calculation, p=7, q=7, and the model is ARIMA (7, 1, 7). The specific results are shown in Table 2 below:

[0094] Table 2 Specific results

[0095]

[0096] The model of standard sensor 1 is established as follows:

[0097]

[0098] The predicted graph obtained is as follows Figure 5 shown.

[0099] pass Figure 5 It can be found that the predicted value and the observed value are basically consistent, and a better model for predicting humidity is obtained. At the same time, the humidity value for the next stage is: 46.0%RH.

[0100] The models of standard sensors 4, 5, 6, 10, and 11 are as follows:

[0101]

[0102]

[0103]

[0104]

[0105]

[0106] As shown in Table 1, the grey comprehensive correlation between the tested sensor and standard sensors 1, 4, 5, 6, 10, and 11 are 0.95, 0.97, 0.99, 0.91, 0.87, and 0.87, respectively. The weighted average coefficients calculated according to expression (9) are 0.17, 0.17, 0.18, 0.16, 0.16, and 0.16.

[0107] The standard mathematical model obtained is:

[0108]

[0109] The standard mathematical model and the waveform diagram of the instrument under test are as follows: Figure 6 As shown, according to the models of standard sensors 1, 4, 5, 6, 10, and 11, the humidity values ​​of the next stage are: 46.0%RH, 45.2%RH, 48.2%RH, 45.0%RH, 44.1%RH, and 47.5%RH. The humidity values ​​of the next stage are brought into the standard mathematical model, and the data of one stage is: 46.0%RH. At this time, the data of the temperature and humidity sensor under test is 47.5%RH. Therefore, the humidity error of the instrument under test is 1.5%RH.

[0110] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have the same meaning as in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined similarly here.

[0111] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A large-scale spatial temperature and humidity assessment method based on multi-information fusion, characterized by: include: The entire space size is evenly divided according to the length and width of the plane size to obtain multiple regional spaces, and the standard sensors are arranged in each regional space; Set the same sampling frequency, record the data of the tested temperature and humidity sensor, collect the data of the standard sensor, obtain the relevant data sequence, and calculate the correlation between the data of the tested temperature and humidity sensor and the data of the standard sensor; According to the calculation result of the correlation, the data of the relevant temperature and humidity sensors under inspection are retained, and the ARIMA model of the time series is established according to the retained data of the temperature and humidity sensors under inspection; Calculating the weighted average coefficient of correlation of the retained data of the inspected temperature and humidity sensors; According to the ARIMA model of the time series and the calculated weighted average coefficient, a standard mathematical model is established; The error of the tested temperature and humidity sensor is inferred based on the standard mathematical model.

2. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 1 is characterized in that: The number of the regional spaces is not less than 9.

3. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 1 is characterized in that: The calculation expression for the correlation between the data of the tested temperature and humidity sensor and the data of the standard sensor is: ； in, is the comprehensive correlation, is the relative correlation, is the absolute correlation, is the correlation coefficient.

4. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 3 is characterized in that: The calculation process of relative correlation is: Compute the initial image of each data sequence: ； ； in, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, is the data sequence of the temperature and humidity sensor under test, is the standard sensor data sequence, It is the first data of the data sequence of the temperature and humidity sensor under test. The first data of the standard sensor data sequence is is the initial image of the temperature and humidity sensor under test data, is the initial image of the standard sensor data, is the number of the tested temperature and humidity sensor data sequence or the standard sensor data sequence; Calculate the distance between data series and : ； ； in, is the spacing of the initial image data sequence of the temperature and humidity sensor under inspection, is the spacing of the initial image data sequence of the standard sensor, is the initial image of the temperature and humidity sensor under inspection, is the initial image of the standard sensor, not subscript The serial number of the data sequence of the temperature and humidity sensor under test or the data sequence of the standard sensor; Calculate the relative correlation: 。 5. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 2 is characterized in that: The standard for retaining the data of the relevant temperature and humidity sensors under inspection is high correlation, which is expressed as: ； in, The comprehensive correlation.

6. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 1 is characterized in that: The expression of the time series ARIMA model is: ； in, is the predicted value, is the autoregressive term coefficient, is the moving average term coefficient, p is the autoregressive term, q is the moving average term, , is the previous value, is the interference term, For time series.

7. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 5 is characterized in that: The calculation expression of the weighted average coefficient of the correlation of the data of the retained temperature and humidity sensors is: ； in, is the weighted average coefficient, is the comprehensive correlation of the data of the detected temperature and humidity sensors that are retained, is the number of standard sensors that are highly correlated with the temperature and humidity sensor data being tested, It is the serial number of the data of the detected temperature and humidity sensor that is retained.

8. The large-scale spatial temperature and humidity assessment method based on multi-information fusion according to claim 1 is characterized in that: The expression of the mathematical model of the standard is: ； in, is the weighted average coefficient, is the predicted value, For the standard model.

Images