An intelligent monitoring method for ambient temperature in pediatric departments

Abstract

The present invention relates to the field of electronic digital data processing technology, and specifically to an intelligent monitoring method for ambient temperature in pediatric departments. The method acquires a temperature signal; based on changes in the temperature signal, obtains an external influence value of an extreme point; uses the difference between the extreme point and the next adjacent extreme point as a change degree value, and obtains an actual skewness based on the distribution of the change degree values ​​and the external influence value; and based on the size of the actual skewness, determines a method for adaptively acquiring Gaussian filter parameters, acquires the Gaussian filter parameters, filters the temperature signal, obtains denoised temperature data, and monitors the ambient temperature data in the pediatric department. The present invention adaptively acquires Gaussian filter parameters, and then accurately denoises the temperature signal using a Gaussian filter algorithm, thereby accurately monitoring the ambient temperature data in the pediatric department.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to an intelligent monitoring method for ambient temperature in a pediatric department. Background Art

[0002] Monitoring ambient temperature in pediatric departments is crucial, especially in neonatal and infant wards. Excessively high or low temperatures can have direct or indirect negative impacts on infants' health. Therefore, accurate, real-time temperature monitoring is crucial to ensuring children's safety and health. Currently, the latest temperature monitoring technology uses the Internet of Things (IoT) to remotely monitor ambient temperature in pediatric departments, enabling automatic recording and remote monitoring of temperature data. However, during the actual temperature data recording process, noise may interfere with the data during collection and transmission, resulting in discrepancies between the recorded and actual temperature data. Therefore, denoising the temperature data is necessary during the monitoring and recording process.

[0003] In the existing method, temperature data is denoised using a Gaussian filtering algorithm. However, the Gaussian filtering algorithm requires manual setting of Gaussian filtering parameters. The manually set Gaussian filtering parameters have errors and cannot accurately denoise the noise generated by different conditions. As a result, the temperature data of the pediatric department environment cannot be accurately monitored, resulting in the inability to accurately control the temperature of the pediatric department environment. Summary of the Invention

[0004] In order to solve the technical problem that the artificially set Gaussian filter parameters in the Gaussian filter algorithm have errors and cannot accurately denoise the noise generated by different conditions, and thus cannot accurately monitor the temperature data of the pediatric department environment, the purpose of the present invention is to provide an intelligent monitoring method for the ambient temperature of the pediatric department. The technical solution adopted is as follows:

[0005] The present invention proposes an intelligent monitoring method for ambient temperature in a pediatric department, which comprises the following steps:

[0006] Obtain temperature data at each moment within a preset time period and convert it into a temperature signal;

[0007] Construct a time series interval for each extreme point on the temperature signal, and obtain the external impact value of each extreme point based on the size and change of each temperature data in each time series interval, as well as the number of temperature data;

[0008] The amplitude difference between each extreme point and the next adjacent extreme point on the temperature signal is used as the change degree value. According to the distribution of each change degree value and the external influence values ​​of the corresponding two extreme points, the actual skewness of the distribution of the change degree values ​​is obtained;

[0009] When the absolute value of the actual skewness is greater than a preset skewness threshold, a change degree value curve is obtained based on the size of the change degree value and the number of values ​​with the same change degree; the change degree value curve is decomposed to obtain a preset number of decomposition curves; the overall energy of each decomposition curve is obtained based on the temperature data and external influence value between two extreme points corresponding to each change degree value in each decomposition curve; the Gaussian filter parameters are obtained based on the overall energy of each decomposition curve and the variance of all corresponding temperature data; when the absolute value of the actual skewness is less than or equal to the preset skewness threshold, the Gaussian filter parameters are obtained based on the fluctuation of the temperature data on the temperature signal;

[0010] The temperature signal is filtered according to the Gaussian filter parameters to obtain the denoised temperature data, and the ambient temperature data of the pediatric department is monitored.

[0011] Furthermore, the method for constructing the time series interval of each extreme point on the temperature signal is:

[0012] The time period consisting of all moments between the i-1th extreme point and the i+1th extreme point on the temperature signal is taken as the time series interval of the i-th extreme point; wherein, the time series interval of the i-th extreme point does not include the moments corresponding to the i-1th extreme point and the i+1th extreme point.

[0013] Furthermore, the calculation formula of the external impact value is:

[0014]

[0015] Where, is the external influence value of the i-th extreme point; is the number of temperature data in the time series interval of the i-th extreme point; is the derivative of the nth temperature data in the time series interval of the i-th extreme point; is the nth temperature data in the time series interval of the i-th extreme point; is the mean of all temperature data in the time series interval of the i-th extreme point; is the absolute value function; tanh is the hyperbolic tangent function; exp is the exponential function with the natural constant e as the base; norm is the normalization function.

[0016] Furthermore, the method for obtaining the actual skewness of the distribution of the change degree values ​​according to the distribution of each change degree value and the external influence values ​​of the corresponding two extreme points is:

[0017] Obtain the ratio of the number of occurrences of each change degree value to the total number of change degree values ​​as the distribution probability of each change degree value;

[0018] Obtaining the skewness of the distribution probability as the skewness of the distribution of the degree of change value;

[0019] The mean of the external influence values ​​of the two extreme points corresponding to each change degree value is used as the correction weight of the corresponding change degree value;

[0020] The skewness of the distribution of the degree of change values ​​is corrected according to the correction weight to obtain the actual skewness of the distribution of the degree of change values.

[0021] Furthermore, the calculation formula for the actual skewness of the distribution of the degree of change value is:

[0022]

[0023] Where, is the actual skewness of the distribution of the degree of change values; J is the total number of degree of change values; is the correction weight of the j-th change value; is the distribution probability of the j-th change degree value; is the mean of the distribution probability of all change degree values; is the standard deviation of the distribution probability of all variation values.

[0024] Furthermore, the method for obtaining the change degree value curve is:

[0025] Arrange the degree of change values ​​from small to large on the horizontal axis, use the number of occurrences of each degree of change value on the horizontal axis as the vertical axis, determine the point corresponding to each degree of change value in the coordinate system, perform curve fitting, and use the obtained curve as the degree of change value curve.

[0026] Furthermore, the method of decomposing the change degree value curve to obtain a preset number of decomposition curves is:

[0027] The decomposition condition in the independent component analysis algorithm is modified to Gaussian, and the change degree value curve is decomposed by the independent component analysis algorithm after the decomposition condition is modified to obtain a preset number of decomposition curves.

[0028] Furthermore, the method for obtaining the overall energy of each decomposition curve according to the temperature data and external influence value between two extreme points corresponding to each change degree value in each decomposition curve is:

[0029] For any change degree value, the time period consisting of the moments corresponding to the two extreme value points corresponding to the change degree value is taken as the target time period;

[0030] Obtaining the energy of the temperature signal segment corresponding to the target time period as the initial energy of the temperature signal segment corresponding to the change degree value;

[0031] Correcting the initial energy according to the correction weight of the change degree value to obtain the actual energy of the temperature signal segment corresponding to the change degree value;

[0032] The result of accumulating the actual energy of the temperature signal segments corresponding to all the change degree values ​​in each decomposition curve is taken as the overall energy of each decomposition curve.

[0033] Furthermore, the calculation formula of the actual energy is:

[0034]

[0035] Where, is the actual energy of the temperature signal segment corresponding to the kth change value; is the correction weight of the kth change degree value; is the initial moment of the temperature signal segment corresponding to the kth change value; is the final moment of the temperature signal segment corresponding to the k-th change value; is the temperature data at the t-th moment in the temperature signal segment corresponding to the k-th change value; is the absolute value symbol.

[0036] Furthermore, the method for obtaining Gaussian filter parameters according to the overall energy of each decomposition curve and the variance of all corresponding temperature data is:

[0037] The result of accumulating the overall energy of all decomposition curves is taken as the first result;

[0038] Obtaining the ratio of the overall energy of each decomposition curve to the first result as the first weight of the corresponding decomposition curve;

[0039] The product of the first weight of each decomposition curve and the variance of all corresponding temperature data is used as the participating variance of the corresponding decomposition curve;

[0040] The result of accumulating the participating variances of all decomposition curves is used as the Gaussian filter parameter.

[0041] The present invention has the following beneficial effects:

[0042] Construct a time series interval for each extreme point on the temperature signal, obtain the external influence value of each extreme point according to the size and change of each temperature data in each time series interval, and the number of temperature data, determine the degree of interference of each extreme point by external influencing factors, denoise the temperature signal under the premise of considering external influencing factors, and reduce the excessive smoothing of the temperature signal; in order to accurately denoise the temperature signal, the amplitude difference between each extreme point and the next adjacent extreme point on the temperature signal is used as the degree of change value, and the complexity of the noise in the temperature signal is analyzed based on the degree of change value, and then the actual skewness of the distribution of the degree of change value is obtained according to the distribution of each degree of change value and the external influence value of the corresponding two extreme points, which accurately reflects the complexity of the noise in the temperature signal. When the absolute value of the actual skewness is less than 0. When the value is greater than the preset skewness threshold, the complexity of the noise in the temperature signal is relatively large. Then, according to the size of the change degree value and the number of the same change degree values, the change degree value curve is obtained to determine the overall distribution of the noise superposition in the temperature signal. The change degree value curve is decomposed to obtain a preset number of decomposition curves, and then the overall energy of each decomposition curve is obtained to distinguish the primary and secondary parts of the noise in the temperature signal. Then, according to the overall energy of each decomposition curve and the variance of all corresponding temperature data, the Gaussian filter parameters are adaptively obtained to avoid incomplete noise removal in the temperature signal; the temperature signal is filtered according to the Gaussian filter parameters to accurately obtain the denoised temperature data, and the environmental temperature data of the pediatric department is accurately monitored to ensure the health and safety of children in the pediatric department. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A flowchart of an intelligent monitoring method for ambient temperature in a pediatric department is provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0045] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following describes in detail, in conjunction with the accompanying drawings and preferred embodiments, a method for intelligently monitoring ambient temperature in a pediatric department, according to the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The following describes in detail a specific solution of an intelligent monitoring method for ambient temperature in a pediatric department provided by the present invention with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a flow chart of an intelligent monitoring method for ambient temperature in a pediatric department provided by one embodiment of the present invention, the method comprising the following steps:

[0049] Step S1: Obtain temperature data at each moment within a preset time period and convert it into a temperature signal.

[0050] Specifically, the embodiment of the present invention obtains and records the temperature data at each moment in the pediatric department through a temperature sensor. In the process of obtaining the temperature data, it may be affected by noise data. In order to monitor the temperature data of the pediatric department environment in real time and accurately and ensure that the temperature data of the pediatric department environment remains normal, it is necessary to denoise the temperature data obtained by the temperature sensor. The embodiment of the present invention obtains the temperature data at each moment within the two hours before the current moment, that is, the preset time period is set to two hours, wherein the last moment of the preset time period is the current moment. At the same time, the time interval between two adjacent moments is set to half a minute. The implementer can set the size of the preset time period and the time interval between two adjacent moments according to actual conditions, which is not limited here. In order to accurately and efficiently denoise the temperature data, the embodiment of the present invention converts the temperature data within the preset time period into a temperature signal, wherein the horizontal axis of the temperature signal is time and the vertical axis is the size of the temperature data.

[0051] The scenario of the embodiment of the present invention is that the temperature sensor has recorded temperature data in the pediatric department for at least two hours before the current moment.

[0052] The purpose of the embodiment of the present invention is: in order to accurately monitor the temperature data in the pediatric department and ensure the safety and health of infants and young children, it is necessary to denoise the temperature data through a Gaussian filtering algorithm. Because the parameters in the Gaussian filtering algorithm need to be set manually, it is easy to cause inaccurate noise removal. Therefore, in order to accurately remove the noise in the temperature data, the embodiment of the present invention analyzes the temperature data within a preset time period, determines the complexity of the noise in the temperature data within the preset time period, and then adaptively obtains the Gaussian filtering parameters in the Gaussian filtering algorithm to ensure that the temperature data within the preset time period is accurately denoised, and then accurately monitors the temperature data of the pediatric department environment, and timely adjusts the temperature in the pediatric department to ensure the safety and health of infants and young children.

[0053] Step S2: Construct a time series interval for each extreme point on the temperature signal, and obtain the external impact value of each extreme point based on the size and change of each temperature data in each time series interval and the number of temperature data.

[0054] Specifically, the temperature data change rate in the pediatric environment is usually low. For the noise signal superimposed on the temperature signal, different local intensities will cause different actual interference to the temperature data. Among them, the external influencing factors in the pediatric environment have a relatively large interference with the temperature data. Therefore, by analyzing the characteristics of the temperature signal itself, the external influence value caused by the external influencing factors can be obtained for each extreme point in the temperature signal. Among them, the external influence refers to other external factors in the pediatric environment that interfere with the measurement accuracy of the temperature sensor. The main external influencing factor is the unstable power supply in the pediatric department. For example, when other electrical medical devices are connected to the circuit of the pediatric department, it will cause the circuit in the pediatric department to have a short-term power supply instability, which will cause the measurement accuracy of the temperature sensor to deviate. The corresponding temperature signal segment will show a large degree of change and a short duration. In order to accurately analyze the degree of influence of the fluctuation in the temperature signal due to unstable power supply, an embodiment of the present invention constructs a time series interval for each extreme point on the temperature signal. The greater the degree of change in the temperature data within the time series interval and the smaller the number of temperature data, that is, the shorter the time series interval, the greater the degree to which the corresponding extreme point is affected by the external environment. Therefore, based on the size and change of each temperature data in each time series interval, as well as the number of temperature data, the external influence value of each extreme point is obtained. Among them, the larger the external influence value, the greater the interference of the corresponding extreme point by external influencing factors. The specific method for obtaining the external influence value of each extreme point is as follows:

[0055] (1) Obtain the time series interval.

[0056] Preferably, the method for constructing the timing interval is: the time period composed of all moments between the i-1th extreme point and the i+1th extreme point on the temperature signal is used as the timing interval of the i-th extreme point; wherein, the timing interval of the i-th extreme point does not include the moments corresponding to the i-1th extreme point and the i+1th extreme point. If the i-th extreme point is the first extreme point on the temperature signal, the timing interval of the i-th extreme point is extended forward to the initial endpoint of the temperature signal; if the i-th extreme point is the last extreme point on the temperature signal, the timing interval of the i-th extreme point is extended backward to the end endpoint of the temperature signal. The implementer can set the size of the timing interval according to the actual situation, which is not limited here. According to the method for obtaining the timing interval of the i-th extreme point, the timing interval of each extreme point on the temperature signal is obtained.

[0057] (2) Obtain external impact value.

[0058] As an example, taking the i-th extreme point as an example, when the i-th extreme point is affected by unstable power supply, the temperature data is significantly affected compared to electromagnetic interference noise, as evidenced by large amplitude fluctuations in the temperature signal segment corresponding to the time series interval of the i-th extreme point. Normal temperature signals corresponding to pediatric departments do not experience significant fluctuations. Therefore, the embodiment of the present invention obtains the variance of the temperature data within the time series interval of the i-th extreme point as the first variance. A larger first variance indicates a greater degree of external influence on the i-th extreme point. Power instability is a change that occurs when a circuit is connected to or disconnected from an electrical appliance. Compared to electromagnetic interference noise, electrical appliances in pediatric departments have a greater impact on power supply. That is, fluctuations in the temperature signal caused by unstable power supply are greater than those caused by electromagnetic interference noise. Therefore, to highlight the difference between unstable power supply and electromagnetic interference noise in the temperature signal, the embodiment of the present invention obtains the derivative of each temperature data point within the time series interval of the i-th extreme point and adjusts the variance of the temperature data within the time series interval of the i-th extreme point. For temperature data with larger derivatives within the time series interval of the i-th extreme point, the probability of being affected by unstable power supply is greater. At the same time, the duration of unstable power supply is shorter. Therefore, when the time series interval of the i-th extreme point is shorter, that is, the number of temperature data within the time series interval of the i-th extreme point is relatively small, it means that the change of the i-th extreme point is more likely to be caused by unstable circuit power supply, resulting in inaccurate measurement accuracy of the temperature sensor. Therefore, based on the first variance and the derivative of each temperature data within the time series interval of the i-th extreme point, as well as the number of temperature data within the time series interval of the i-th extreme point, the calculation formula for obtaining the external influence value of the i-th extreme point is:

[0059]

[0060] Where, is the external influence value of the i-th extreme point; is the number of temperature data in the time series interval of the i-th extreme point; is the derivative of the nth temperature data in the time series interval of the i-th extreme point; is the nth temperature data in the time series interval of the i-th extreme point; is the mean of all temperature data in the time series interval of the i-th extreme point; is the absolute value function; tanh is the hyperbolic tangent function; exp is the exponential function with the natural constant e as the base; norm is the normalization function.

[0061] It should be noted that The smaller it is, the shorter the time interval of the i-th extreme point is, and the i-th extreme point is more likely to be affected by power supply instability factors. The bigger, The bigger; The larger the value, the more unstable the temperature data in the time series interval of the i-th extreme point is. right Make corrections to accurately determine the degree of external influence on the i-th extreme point. The larger the value is, the more likely the change in temperature data within the time series interval of the i-th extreme point is to be caused by unstable power supply. The larger; therefore, The larger the value is, the greater the fluctuation of the i-th extreme point is affected by the unstable power supply factor. The value range of is 0 to 1. According to the method for obtaining the external influence value of the i-th extreme point, the external influence value of each extreme point on the temperature signal is obtained.

[0062] Step S3: The amplitude difference between each extreme point and the next adjacent extreme point on the temperature signal is used as the change degree value, and the actual skewness of the change degree value distribution is obtained according to the distribution of each change degree value and the external influence values ​​of the corresponding two extreme points.

[0063] Specifically, various medical devices are widely used in pediatric departments, and the electromagnetic interference noise to which the temperature signal is subjected is relatively complex. The electromagnetic interference noise conforms to a Gaussian distribution. In order to prevent the Gaussian filter algorithm from over-smoothing the temperature signal or the denoising intensity from being too low, it is necessary to further analyze the temperature signal. By analyzing the complexity of the noise in the temperature signal, the Gaussian filter parameters are adaptively obtained to accurately denoise the temperature signal. In order to accurately analyze the complexity of the noise in the temperature signal, an embodiment of the present invention obtains the absolute value of the difference between the amplitude of each extreme point on the temperature signal and the next adjacent extreme point as the degree of change value. For the last extreme point on the temperature signal, the absolute value of the difference between the amplitude of the last extreme point and the previous adjacent extreme point is used as the degree of change value of the last extreme point. When the number of noise types in the temperature signal is smaller, the distribution of the degree of variation values ​​is more stable. When the number of noise types in the temperature signal is greater, i.e., the noise is more complex, the distribution of the degree of variation values ​​is less stable. Therefore, embodiments of the present invention obtain the distribution probability of each degree of variation value based on the distribution of the degree of variation values, and further obtain the skewness of the distribution probability. The closer the skewness is to 0, the fewer the noise types in the temperature signal are; the further the skewness is from 0, the more the noise types in the temperature signal are. The method for obtaining the skewness is a well-known calculation and will not be described in detail here. In actual situations, fluctuations in the temperature signal may be caused by unstable power supply. Therefore, the degree of variation value may also be caused by unstable power supply. To more accurately analyze the complexity of the noise in the temperature signal and reduce the interference of unstable power supply on noise type detection, embodiments of the present invention correct the skewness of the distribution probability based on the external influence value of each extreme point to obtain the actual skewness, which is used to represent the noise complexity in the temperature signal. Therefore, based on the distribution of each degree of variation value and the external influence values ​​of the corresponding two extreme points, the actual skewness of the degree of variation value distribution is obtained. The larger the actual skewness, the more types of electromagnetic interference noise in the temperature signal.

[0064] Preferably, the method for obtaining the actual skewness is: obtaining the ratio of the number of occurrences of each degree of change value to the total number of degree of change values ​​as the distribution probability of each degree of change value; obtaining the skewness of the distribution probability as the skewness of the distribution of the degree of change values; the larger the skewness, the more unstable the distribution probability, which indirectly indicates that there are more types of noise in the temperature signal; in order to reduce the impact of unstable power supply on the skewness, the average of the external influence values ​​of the two extreme points corresponding to each degree of change value is used as the correction weight of the corresponding degree of change value; the skewness of the distribution of the degree of change values ​​is corrected according to the correction weight to obtain the actual skewness of the distribution of the degree of change values.

[0065] As an example, the calculation formula for obtaining the actual skewness of the distribution of degree of change values ​​is:

[0066]

[0067] Where, is the actual skewness of the distribution of the degree of change values; J is the total number of degree of change values; is the correction weight of the j-th change value; is the distribution probability of the j-th change degree value; is the mean of the distribution probability of all change degree values; is the standard deviation of the distribution probability of all variation values.

[0068] It should be noted that through right Make corrections so that More accurately represents the complexity of the noise in the temperature signal, The larger the value is, the greater the external influence on the two extreme points corresponding to the j-th change degree value is, and the smaller the weight of the distribution probability of the j-th change degree value in calculating the skewness should be. Therefore, through right Perform negative correlation processing. The smaller it is, the less the j-th change value participates in obtaining the skewness. The smaller it is, the fewer types of noise there are in the temperature signal.

[0069] Step S4: When the absolute value of the actual skewness is greater than a preset skewness threshold, a change degree value curve is obtained according to the size of the change degree value and the number of values ​​with the same change degree; the change degree value curve is decomposed to obtain a preset number of decomposition curves; the overall energy of each decomposition curve is obtained according to the temperature data and external influence value between two extreme points corresponding to each change degree value in each decomposition curve; the Gaussian filter parameter is obtained according to the overall energy of each decomposition curve and the variance of all corresponding temperature data; when the absolute value of the actual skewness is less than or equal to the preset skewness threshold, the Gaussian filter parameter is obtained according to the fluctuation of the temperature data on the temperature signal.

[0070] Specifically, step S3 shows that the actual skewness represents the complexity of the superposition of different types of noise in the temperature signal. A larger actual skewness indicates a greater number of noise types in the temperature signal. To achieve optimal denoising of temperature data, this embodiment of the present invention sets a preset skewness threshold of 3. Implementers can adjust the preset skewness threshold based on actual conditions, and this is not limited here. When the absolute value of the actual skewness is greater than the preset skewness threshold, the temperature signal is subject to significant electromagnetic interference noise. To avoid over-smoothing the temperature signal or insufficient denoising, this embodiment of the present invention adaptively obtains Gaussian filter parameters based on the degree of variation value. Because electromagnetic interference noise in the temperature signal exhibits Gaussian properties, the degree of variation value is essentially obtained by superimposing different types of electromagnetic interference noise. When Gaussian noise is superimposed, the resulting degree of variation value also exhibits Gaussian properties. Therefore, the degree of variation values ​​are arranged from smallest to largest on the horizontal axis, with the number of occurrences of each degree of variation value on the horizontal axis serving as the vertical axis. The points corresponding to each degree of variation value in the coordinate system are determined, and curve fitting is performed. The resulting curve is used as the degree of variation curve. Curve fitting is a conventional technique and will not be further described. The variation value curve is actually the superposition result of the same variation interval and the superposition result of different variation intervals of different electromagnetic interference noise. Therefore, the embodiment of the present invention sets the preset number to 2. The implementer can set the size of the preset number according to actual conditions, and it is not limited here. That is, the variation value curve is decomposed into two decomposition curves. The decomposition curve is the curve corresponding to the superposition of electromagnetic interference noise. Therefore, when the embodiment of the present invention decomposes the variation value curve using the independent component analysis algorithm, it is necessary to modify the decomposition condition in the independent component analysis algorithm from non-Gaussian to Gaussian. The independent component analysis algorithm is a well-known technology and will not be described in detail. In order to adaptively obtain the Gaussian filter parameters, the overall energy of each decomposition curve is obtained based on the temperature data and external influence value between the two extreme points corresponding to each variation value in each decomposition curve. The larger the overall energy, the more dominant the noise in the corresponding decomposition curve.

[0071] Preferably, the method for obtaining the overall energy is: for any change degree value, the time period formed by the moments corresponding to the two extreme points corresponding to the change degree value is used as the target time period; the energy of the temperature signal segment corresponding to the target time period is obtained as the initial energy of the temperature signal segment corresponding to the change degree value; due to external influencing factors, that is, the influence of unstable power supply on the temperature signal amplitude is usually greater than the influence of electromagnetic interference noise on the temperature signal, therefore, in order to reduce the interference of external influencing factors, the initial energy is corrected according to the correction weight of the change degree value, and the actual energy of the temperature signal segment corresponding to the change degree value is obtained; the larger the actual energy, the greater the proportion of the change degree value in obtaining the overall energy of the decomposition curve; the result of accumulating the actual energies of the temperature signal segments corresponding to all change degree values ​​in each decomposition curve is used as the overall energy of each decomposition curve. Among them, the signal energy acquisition method is a well-known technology and will not be described in detail.

[0072] As an example, taking the kth degree of change value as an example, the kth degree of change value is obtained by the absolute value of the difference between the amplitudes of two adjacent extreme points. The two extreme points corresponding to the kth degree of change value are respectively taken as the kth extreme point and the k+1th extreme point. The time period formed by the moments corresponding to the kth extreme point and the k+1th extreme point is the target time period corresponding to the kth degree of change value. Obtain the energy in the temperature signal segment corresponding to the target time period, and determine the proportion of temperature data in the target time period in the overall temperature signal. The calculation formula for obtaining the actual energy of the temperature signal segment corresponding to the kth degree of change value is:

[0073]

[0074] Where, is the actual energy of the temperature signal segment corresponding to the kth change value; is the correction weight of the kth change degree value; is the initial moment of the temperature signal segment corresponding to the kth change value; is the final moment of the temperature signal segment corresponding to the k-th change value; is the temperature data at the t-th moment in the temperature signal segment corresponding to the k-th change value; is the absolute value symbol.

[0075] It should be noted that To integrate and sum the temperature data within the target time period corresponding to the k-th change degree value, that is, the initial energy of the temperature signal segment corresponding to the k-th change degree value, through right Make corrections to reduce the interference of external factors, so that The more accurate. The larger the k-th change degree value is, the greater the proportion of electromagnetic interference noise in the temperature signal segment within the target time period.

[0076] The actual energy of each temperature signal segment corresponding to each degree of change value is obtained using the method for obtaining the actual energy of the temperature signal segment corresponding to the kth degree of change value. The actual energy of the temperature signal segments corresponding to all degree of change values ​​within each decomposition curve is accumulated as the overall energy of each decomposition curve.

[0077] For temperature signals with less electromagnetic interference noise, a relatively ideal denoising effect can be obtained by adjusting the Gaussian filter parameters in the Gaussian filter algorithm. Therefore, the embodiment of the present invention obtains the variance of all temperature data corresponding to each decomposition curve as the main part of the Gaussian filter parameter. The variance of the temperature data corresponding to each decomposition curve is adjusted according to the proportion of the overall energy of each decomposition curve, and the Gaussian filter parameter is adaptively obtained.

[0078] Preferably, the method for obtaining Gaussian filter parameters is: taking the result of accumulating the overall energy of all decomposition curves as the first result; obtaining the ratio of the overall energy of each decomposition curve to the first result as the first weight of the corresponding decomposition curve; multiplying the first weight of each decomposition curve by the variance of all corresponding temperature data as the participating variance of the corresponding decomposition curve; and taking the result of accumulating the participating variance of all decomposition curves as the Gaussian filter parameter.

[0079] As an example, the calculation formula for obtaining Gaussian filter parameters is:

[0080]

[0081] Where, is the Gaussian filter parameter; is the overall energy of the first decomposition curve; is the overall energy of the second decomposition curve; is the variance of all temperature data corresponding to the first decomposition curve; is the variance of all temperature data corresponding to the second decomposition curve; is the first result; is the first weight of the first decomposition curve; is the first weight of the second decomposition curve; is the participation variance of the first decomposition curve; is the participation variance of the second decomposition curve.

[0082] It should be noted that The larger it is, the more likely the noise in the first decomposition curve is to be the main noise in the temperature signal, and the more likely it is to be the main noise in the temperature signal. The greater the contribution rate to obtaining Gaussian filter parameters; The smaller it is, the more likely the noise in the second decomposition curve is the secondary noise in the temperature signal. The smaller the contribution rate to obtaining Gaussian filter parameters.

[0083] When the absolute value of the actual skewness is less than or equal to the preset skewness threshold, it means that the electromagnetic interference noise superimposed on the temperature signal is relatively small, that is, the complexity of the electromagnetic interference noise is low, and the temperature data in the corresponding pediatric department changes smoothly. Therefore, the variance of all temperature data on the temperature signal is obtained as the Gaussian filter parameter.

[0084] Step S5: Filter the temperature signal according to the Gaussian filter parameters to obtain denoised temperature data, and monitor the ambient temperature data of the pediatric department.

[0085] The Gaussian filter parameter is the variance in the Gaussian filter algorithm. The temperature signal is then filtered using the Gaussian filter algorithm based on the Gaussian filter parameter to obtain denoised temperature data. If the denoised temperature data is within the set temperature range within a preset time period, it indicates that the current temperature setting in the pediatric department is reasonable. If the denoised temperature data is not within the set temperature range within the preset time period, it indicates that the current temperature setting in the pediatric department is unreasonable. The temperature control device will issue an alarm to remind staff to adjust the temperature data in the pediatric department in a timely manner to ensure that the temperature in the pediatric department is within the set temperature range to ensure the health and safety of infants and young children. The set temperature range should be based on the actual date, time, and temperature requirements. For example, during summer noon, the temperature range in the pediatric department environment should be set between 22° and 26° to ensure that the temperature in the pediatric department is not too hot and prevent excessive sweating in infants and young children. During summer nights, the temperature range in the pediatric department environment should be set between 18° and 22° to ensure that the temperature in the pediatric department is not too cold and prevent infants and young children from catching colds.

[0086] So far, the present invention is completed.

[0087] In summary, the embodiment of the present invention obtains a temperature signal; obtains the external influence value of the extreme point based on the change of the temperature signal; uses the difference between the extreme point and the next adjacent extreme point as the change degree value, and obtains the actual skewness based on the distribution of the change degree value and the external influence value; determines the method of adaptively obtaining Gaussian filter parameters based on the size of the actual skewness, obtains the Gaussian filter parameters to filter the temperature signal, obtains denoised temperature data, and monitors the ambient temperature data of the pediatric department. The present invention adaptively obtains Gaussian filter parameters, and then accurately denoises the temperature signal through the Gaussian filter algorithm, thereby accurately monitoring the ambient temperature data of the pediatric department.

[0088] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent monitoring method for ambient temperature in pediatric departments, characterized in that: The method comprises the following steps: Obtain temperature data at each moment within a preset time period and convert it into a temperature signal; Construct a time series interval for each extreme point on the temperature signal, and obtain the external impact value of each extreme point based on the size and change of each temperature data in each time series interval, as well as the number of temperature data; The amplitude difference between each extreme point and the next adjacent extreme point on the temperature signal is used as the change degree value. According to the distribution of each change degree value and the external influence values ​​of the corresponding two extreme points, the actual skewness of the distribution of the change degree values ​​is obtained; When the absolute value of the actual skewness is greater than a preset skewness threshold, a change degree value curve is obtained based on the size of the change degree value and the number of values ​​with the same change degree; the change degree value curve is decomposed to obtain a preset number of decomposition curves; the overall energy of each decomposition curve is obtained based on the temperature data and external influence value between two extreme points corresponding to each change degree value in each decomposition curve; the Gaussian filter parameters are obtained based on the overall energy of each decomposition curve and the variance of all corresponding temperature data; when the absolute value of the actual skewness is less than or equal to the preset skewness threshold, the Gaussian filter parameters are obtained based on the fluctuation of the temperature data on the temperature signal; The temperature signal is filtered according to the Gaussian filter parameters to obtain the denoised temperature data, and the ambient temperature data of the pediatric department is monitored.

2. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The method for constructing the time series interval of each extreme point on the temperature signal is: The time period consisting of all moments between the i-1th extreme point and the i+1th extreme point on the temperature signal is taken as the time series interval of the i-th extreme point; wherein, the time series interval of the i-th extreme point does not include the moments corresponding to the i-1th extreme point and the i+1th extreme point.

3. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The calculation formula of the external impact value is: Where, is the external influence value of the i-th extreme point; is the number of temperature data in the time series interval of the i-th extreme point; is the derivative of the nth temperature data in the time series interval of the i-th extreme point; is the nth temperature data in the time series interval of the i-th extreme point; is the mean of all temperature data in the time series interval of the i-th extreme point; is the absolute value function; tanh is the hyperbolic tangent function; exp is the exponential function with the natural constant e as the base; norm is the normalization function.

4. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The method for obtaining the actual skewness of the distribution of the degree of change values ​​according to the distribution of each degree of change value and the external influence values ​​of the corresponding two extreme points is: Obtain the ratio of the number of occurrences of each change degree value to the total number of change degree values ​​as the distribution probability of each change degree value; Obtaining the skewness of the distribution probability as the skewness of the distribution of the degree of change value; The mean of the external influence values ​​of the two extreme points corresponding to each change degree value is used as the correction weight of the corresponding change degree value; The skewness of the distribution of the degree of change values ​​is corrected according to the correction weight to obtain the actual skewness of the distribution of the degree of change values.

5. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 4, characterized in that: The calculation formula for the actual skewness of the distribution of the degree of change value is: Where, is the actual skewness of the distribution of the degree of change values; J is the total number of degree of change values; is the correction weight of the j-th change value; is the distribution probability of the j-th change degree value; is the mean of the distribution probability of all change degree values; is the standard deviation of the distribution probability of all variation values.

6. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The method for obtaining the change degree value curve is: Arrange the degree of change values ​​from small to large on the horizontal axis, use the number of occurrences of each degree of change value on the horizontal axis as the vertical axis, determine the point corresponding to each degree of change value in the coordinate system, perform curve fitting, and use the obtained curve as the degree of change value curve.

7. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The method of decomposing the change degree value curve to obtain a preset number of decomposition curves is: The decomposition condition in the independent component analysis algorithm is modified to Gaussian, and the change degree value curve is decomposed by the independent component analysis algorithm after the decomposition condition is modified to obtain a preset number of decomposition curves.

8. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 4, characterized in that: The method for obtaining the overall energy of each decomposition curve according to the temperature data and external influence value between two extreme points corresponding to each change degree value in each decomposition curve is: For any change degree value, the time period consisting of the moments corresponding to the two extreme value points corresponding to the change degree value is taken as the target time period; Obtaining the energy of the temperature signal segment corresponding to the target time period as the initial energy of the temperature signal segment corresponding to the change degree value; Correcting the initial energy according to the correction weight of the change degree value to obtain the actual energy of the temperature signal segment corresponding to the change degree value; The result of accumulating the actual energy of the temperature signal segments corresponding to all the change degree values ​​in each decomposition curve is taken as the overall energy of each decomposition curve.

9. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 8, characterized in that: The calculation formula of the actual energy is: Where, is the actual energy of the temperature signal segment corresponding to the kth change value; is the correction weight of the kth change degree value; is the initial moment of the temperature signal segment corresponding to the kth change value; is the final moment of the temperature signal segment corresponding to the k-th change value; is the temperature data at the t-th moment in the temperature signal segment corresponding to the k-th change value; is the absolute value symbol.

10. The method for intelligently monitoring ambient temperature in a pediatric department according to claim 1, wherein: The method for obtaining Gaussian filter parameters based on the overall energy of each decomposition curve and the variance of all corresponding temperature data is as follows: The result of accumulating the overall energy of all decomposition curves is taken as the first result; Obtaining the ratio of the overall energy of each decomposition curve to the first result as the first weight of the corresponding decomposition curve; The product of the first weight of each decomposition curve and the variance of all corresponding temperature data is used as the participating variance of the corresponding decomposition curve; The result of accumulating the participating variances of all decomposition curves is used as the Gaussian filter parameter.

Images