WO2017185796A1 - 一种温度预测方法及电子体温计 - Google Patents

一种温度预测方法及电子体温计 Download PDF

Info

Publication number
WO2017185796A1
WO2017185796A1 PCT/CN2016/113112 CN2016113112W WO2017185796A1 WO 2017185796 A1 WO2017185796 A1 WO 2017185796A1 CN 2016113112 W CN2016113112 W CN 2016113112W WO 2017185796 A1 WO2017185796 A1 WO 2017185796A1
Authority
WO
WIPO (PCT)
Prior art keywords
temperature
value
sampling time
current sampling
temperature measurement
Prior art date
Application number
PCT/CN2016/113112
Other languages
English (en)
French (fr)
Inventor
赵巍
Original Assignee
广州视源电子科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 广州视源电子科技股份有限公司 filed Critical 广州视源电子科技股份有限公司
Publication of WO2017185796A1 publication Critical patent/WO2017185796A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
    • G16Z99/00—Subject matter not provided for in other main groups of this subclass

Definitions

  • the invention relates to the field of electronic thermometers, in particular to a temperature prediction method and an electronic thermometer.
  • the electronic thermometer Compared with the traditional mercury glass thermometer, the electronic thermometer has the advantages of convenient reading, harmless to the human body and the surrounding environment (no mercury), and is suitable for home use.
  • the temperature probe of the electronic thermometer due to the size limitation of the temperature probe of the electronic thermometer, the movement of the human body and the way the user wears it often takes a long time to reach a steady state (heat balance), that is, it takes a long time to measure the true temperature of the object.
  • the general electronic thermometer also uses a prediction algorithm to predict the true temperature of the measured object in advance to improve the measurement speed of the electronic thermometer.
  • the existing prediction algorithms mostly use the method based on the function model to calculate the temperature.
  • the prediction first uses the historical data of a certain length and the predetermined curve type (such as logarithmic curve, hyperbola, etc.) to calculate the parameters of the function model.
  • the value of the function model at a certain point in time is then output as the true temperature of the measured object.
  • the function model includes a first-order function model and a high-order function model.
  • the first-order function model has limited expression ability, and the high-order function model easily leads to over-fitting, so after training the function model,
  • the parameters of the function model must be optimized.
  • the optimization of the current function model is mostly based on empirical parameters, and the scope of application is limited, which cannot meet the requirements of use.
  • the present invention provides a temperature prediction method and an electronic thermometer, which predicts the measured temperature by setting an appropriate first offset, thereby ensuring fast and accurate prediction and wide application range.
  • the invention provides a temperature prediction method, comprising the following steps:
  • n is an integer greater than 2;
  • the temperature prediction value at the current sampling time is output.
  • the method further comprises:
  • the temperature measurement value measured at the current sampling time is output.
  • the method further comprises:
  • the start temperature prediction is stopped, and the temperature measurement value measured at the current sampling time is output;
  • the stopping start condition is that detecting that the continuous start time of the temperature prediction reaches a preset first time threshold or detecting that the continuous start time of the temperature prediction reaches a preset second time threshold, and the current sampling time And the difference between the maximum value and the minimum value of the temperature measurement values of the first t-1 sampling moments of the current sampling time is less than a preset threshold.
  • the method further includes:
  • the method further comprises:
  • the calculating the first offset includes:
  • the initial conditions include the target temperature of the measured object;
  • the invention also provides an electronic thermometer comprising:
  • a temperature acquiring unit configured to acquire a temperature measurement value measured at a current sampling time
  • a first determining unit configured to: after detecting that the temperature prediction has been started, and not satisfying the preset stop start condition, determining whether there is a stable flag; if yes, notifying the first setting unit; if not, notifying the first extraction unit;
  • the first setting unit is configured to set a temperature prediction value of a current sampling time as a temperature prediction value of a previous sampling time;
  • the first extracting unit is configured to acquire temperature measurement values of the first sampling time and the first sampling time of the current sampling time, and extract a maximum value and a minimum value of the n temperature measurement values, where n is An integer greater than 2;
  • a second determining unit configured to determine whether a difference between the maximum value and the minimum value of the n temperature measurement values is greater than a preset threshold; if yes, notify the second setting unit, and if not, notify the third setting unit;
  • a second setting unit configured to generate a temperature prediction value of the current sampling moment according to the temperature measurement value of the current sampling moment and the pre-stored first offset amount
  • a third setting unit configured to set a temperature prediction value of the current sampling time as a temperature prediction value of the last sampling time, and generate the stable flag
  • An output unit for outputting a temperature prediction value at a current sampling time.
  • the electronic thermometer further includes a third determining unit and a starting unit, wherein:
  • the third determining unit is configured to: according to the temperature measurement value of the current sampling time and the temperature measurement value of at least the previous sampling time, when determining the unstarted temperature prediction, and determine whether the preset starting condition is met; if yes, Notifying the startup unit, if not, notifying the output unit;
  • the starting unit is configured to start temperature prediction
  • the output unit is further configured to output a temperature measurement value measured at a current sampling time.
  • the electronic thermometer further comprises:
  • Stopping the starting unit after stopping the preset stopping start condition, stopping the starting temperature prediction, and notifying the output unit;
  • the output unit is further configured to output a temperature measurement value measured at a current sampling time
  • the stopping start condition is that detecting that the continuous start time of the temperature prediction reaches a preset first time threshold or detecting that the continuous start time of the temperature prediction reaches a preset second time threshold, and the current sampling time And the difference between the maximum value and the minimum value of the temperature measurement values of the first t-1 sampling moments of the current sampling time is less than a preset threshold.
  • the electronic thermometer further comprises:
  • a second extraction unit configured to acquire temperature measurement values of the first m-1 sampling moments of the current sampling time and the current sampling time, and extract a maximum value and a minimum value of the m temperature measurement values, where m is greater than 2.
  • a correction unit configured to: when the difference between the maximum value and the minimum value of the m temperature measurement values is greater than the threshold value, correct the temperature prediction value of the current sampling time to a temperature measurement value of the current sampling time plus a second offset preset on the second offset; wherein the second offset is less than or equal to the first offset.
  • the method further comprises:
  • a calculating unit configured to calculate a first offset
  • the computing unit specifically includes:
  • a slope calculation module configured to calculate a slope of the temperature curve within a predetermined time window according to the n and the threshold;
  • a temperature value calculation module configured to derivate the temperature curve and obtain a predetermined x value according to the slope, and obtain a y value corresponding to the x value according to the x value;
  • the first offset calculation module is configured to calculate a first offset according to the target temperature and the y value.
  • the temperature prediction method and the electronic thermometer provided by the embodiments of the present invention only need to add a suitable first offset to the temperature measurement when performing temperature prediction, so the required calculation amount is small, the calculation speed is fast, and the saving is saved.
  • FIG. 1 is a schematic flow chart of a temperature prediction method according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of a comparison between a temperature measurement curve and a temperature prediction value curve provided by an embodiment of the present invention.
  • FIG. 3 is a schematic flow chart of a temperature prediction method according to another embodiment of the present invention.
  • thermometer 4 is a schematic structural view of an electronic thermometer according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of an electronic thermometer according to another embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of an electronic thermometer according to another embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of an electronic thermometer according to another embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of an electronic thermometer according to another embodiment of the present invention.
  • FIG. 9 is a schematic structural view of the computing unit shown in FIG. 8.
  • an embodiment of the present invention provides a temperature prediction method, which can be performed by an electronic thermometer for predicting the temperature of a measured object (such as a human body) when using an electronic thermometer for temperature measurement, including the following step:
  • the electronic thermometer can collect the temperature measurement value of the current sampling time by using the temperature sensing probe, wherein the temperature sensing probe can be a thermal resistance probe, a thermocouple probe, etc., which is not specifically limited in the present invention. .
  • the temperature sensing probe collects a temperature measurement value every predetermined time, for example, collecting one temperature measurement value per second.
  • the specific collection frequency can be set according to actual needs, and the present invention does not Specifically limited.
  • the electronic thermometer in order to obtain the real temperature of the measured object relatively quickly, the electronic thermometer needs to start the temperature prediction at a predetermined stage of the detection (such as a period between a rapid temperature rising phase and reaching the thermal equilibrium), According to the historical temperature measurement value, the temperature prediction value closer to the real temperature of the measured object is generated, and once the temperature sensing probe reaches the thermal equilibrium, there is no need to make prediction, that is, after the heat balance is reached, the startup temperature prediction can be stopped, and the direct output can be directly output. Temperature measurement measured by the temperature probe.
  • the operation of the electronic thermometer in the temperature prediction phase i.e., the operation of the stage in which the temperature prediction is started but the heat balance has not yet been reached) will be specifically described below.
  • the electronic thermometer when the electronic thermometer detects that the temperature prediction has been started and does not satisfy the preset stop start condition, it first determines whether there is a stable mark in the memory.
  • the electronic thermometer can obtain the temperature measurement values of the first sampling time of the current sampling time and the current sampling time, and extract the 20 Temperature measurement The maximum and minimum values in .
  • S105 Determine whether a difference between the maximum value and the minimum value of the n temperature measurement values is greater than a preset threshold.
  • the threshold may be set to 0.05. It should be understood that, in other embodiments of the present invention, the threshold may be set according to actual needs, which is not specifically limited by the present invention.
  • the difference between the maximum value and the minimum value of the n temperature measurement values is greater than the threshold value, it indicates that the temperature measurement value measured by the temperature sensing probe still has a large fluctuation, that is, has not entered yet.
  • it is necessary to generate a temperature prediction value of the current sampling time according to the temperature measurement value of the current sampling time and the pre-stored first offset amount, wherein the temperature prediction value the temperature measurement value+the first offset amount.
  • the temperature prediction value at the current sampling time is set as the temperature prediction value of the previous sampling time, and the stable flag is generated.
  • the difference between the maximum value and the minimum value of the n temperature measurement values is less than the threshold value, it can be considered that the temperature measurement value measured by the temperature sensing probe is about to reach a stable state (ie, the heat balance is about to be reached), so The electronic thermometer maintains the predicted temperature value and generates a stable mark for marking that the temperature probe is about to reach stability. Thereafter, the temperature prediction value at each sampling instant is set to the temperature prediction value of the previous sampling timing.
  • x i is the temperature measurement value of the current sampling time
  • max(x) is the maximum value of the temperature measurement values of n sampling times
  • min(x) is the minimum value of the temperature measurement values of n sampling times
  • y i is the temperature prediction value of the current time
  • o 1 is the first offset
  • ⁇ is the preset threshold
  • flag is the stable flag.
  • the temperature prediction method provided by the embodiment of the present invention can obtain the temperature prediction value by adding the appropriate first offset amount to the temperature measurement value when performing temperature prediction, so the required calculation amount is very Less, the calculation speed is fast, and a lot of computing resources are saved.
  • the principle of temperature prediction is only to add a suitable first offset to the temperature measurement value, this limits the upper limit of the error in principle. With an offset, the prediction error is limited, and there is no case where a large error occurs due to overfitting.
  • step S103 after step S103, before step S108, the method further includes:
  • the electronic thermometer may consider that the temperature probe is about to reach stability in advance, so the stable mark is generated in advance when the stability is not actually achieved. In order to avoid this, the electronic thermometer also sets a condition for returning to the prediction, that is, the predicted temperature remains unchanged, but if the magnitude of the change in the temperature measurement within a certain time range is greater than the threshold (ie, m). The difference between the maximum value and the minimum value in the temperature measurement value is greater than the threshold value, for example, the variation range of the temperature measurement value acquired within 10 seconds exceeds 0.05 degrees), and the temperature prediction value at the current sampling time is re-predicted.
  • the second offset may be equal to the first offset or another empirical value (eg, the second offset may be 0.3).
  • the predicted temperature profile is also smoothed when the temperature prediction is started and the start temperature prediction is stopped.
  • the first step can be gradually increased.
  • the value of the offset is used to achieve a smoothing effect, wherein the first offset can be gradually increased in a linear manner, which is not specifically limited in the present invention.
  • the first offset gradually increases to 0.7 degrees when the temperature prediction is just started, rather than a mutation.
  • the second offset value was 0.3 degrees
  • Correct temperature stability In the embodiment of the present invention, the temperature prediction value is corrected by setting the recovery prediction condition, so that the temperature prediction of the present invention has strong anti-interference ability, and in the case of interference, the temperature prediction value changes with the temperature measurement value. However, when the disturbance decreases or disappears and the fluctuation of the temperature measurement value is small, the temperature prediction value will reach stability again.
  • the electronic thermometer obtains the temperature prediction value by adding the first offset to the temperature measurement, the selection of the first offset is accurate to the prediction. And the speed of prediction has a significant impact.
  • the process of selecting the first offset by the embodiment of the present invention will be described in detail below. specifically:
  • the temperature prediction method further includes:
  • the initial conditions include the target temperature of the measured object.
  • x time (s)
  • y temperature
  • a, b curve parameters. Because the electronic thermometer usually reaches stable at around 15min (900s), that is, the temperature rises no more than 0.15 degrees in 5min (the body temperature does not exceed 37.15 degrees at 1200s). Let the target temperature be 37 degrees of normal body temperature, so you can build a system of equations.
  • the temperature rise curve of the electronic thermometer is a logarithmic curve, in a short time window, the temperature rise curve can be approximated by a straight line. If a temperature rise of 0.05 degrees ( ⁇ y, that is, the threshold) is selected as the judgment condition for predicting whether the temperature is started within 20 s ( ⁇ t, that is, the time required for n sampling times), the slope in the time window is:
  • the first offset can be slightly reduced to 0.7 degrees in consideration of the error between the theoretical curve and the actual measurement.
  • FIG. 3 is a schematic flowchart diagram of a temperature prediction method according to another embodiment of the present invention. It includes at least the following steps:
  • the stop start condition is that the continuous start time of the temperature prediction is detected to reach a preset first time threshold (eg, the start time of the temperature prediction exceeds 1 hour) or the continuous start time of the temperature prediction is detected to be pre- a second time threshold is set, and a difference between a maximum value and a minimum value of the temperature measurement values of the current sampling time and the first t-1 sampling time of the current sampling time is less than a preset threshold, for example, the temperature prediction starts for more than 20 minutes, and The magnitude of the change in temperature measurement over a period of time is less than the threshold (eg, the magnitude of the change in temperature measurement within 10 seconds is less than 0.05 degrees).
  • a preset first time threshold eg, the start time of the temperature prediction exceeds 1 hour
  • a second time threshold is set, and a difference between a maximum value and a minimum value of the temperature measurement values of the current sampling time and the first t-1 sampling time of the current sampling time is less than a preset threshold, for example, the temperature prediction starts for more
  • S204 Determine, according to the temperature measurement value of the current sampling time and the temperature measurement value of at least the previous sampling time, whether the preset starting condition is reached; if yes, go to step S208; if no, execute S205.
  • the temperature prediction is started. (1) The temperature measurement is higher than 30 degrees; (2) the first derivative of the temperature measurement is less than 0.07; (3) the second derivative of the temperature measurement is less than zero.
  • the temperature measurement value of the current sampling time is x i
  • the temperature measurement value of the previous sampling time of the current sampling time is x i-1
  • the temperature measurement value of the first two sampling moments is x i-2
  • the current sampling First derivative of time x i ' x i -x i-1
  • the temperature prediction value of the current sampling time is set as the temperature prediction value of the previous sampling time
  • S208 Acquire a temperature measurement value of a current sampling time and a first sampling time of the current sampling time, and extract a maximum value and a minimum value of the n temperature measurement values, where n is an integer greater than 2.
  • S210 Generate a temperature prediction value of the current sampling time according to the temperature measurement value of the current sampling time and the pre-stored first offset amount.
  • the temperature prediction value of the current sampling time is set as the temperature prediction value of the previous sampling time, and the stable flag is generated.
  • the temperature prediction method provided by the embodiment of the invention provides a start condition for entering the temperature prediction and a stop start condition for the temperature prediction, and in the temperature prediction, the temperature measurement value only needs to be added with a suitable first offset to generate The temperature is predicted, so the amount of calculation required is small, the calculation speed is fast, and a lot of computing resources are saved.
  • the principle of prediction is only to add a suitable first offset to the temperature measurement, this limits the upper limit of the error to the first offset in principle, and the prediction error is limited, and it does not occur because A situation in which a large error occurs due to fitting.
  • FIG. 4 is a schematic structural diagram of an electronic thermometer according to an embodiment of the present invention.
  • the electronic thermometer 100 can be used to perform the temperature prediction method described above, including:
  • the temperature acquiring unit 10 is configured to acquire a temperature measurement value measured at a current sampling time.
  • the temperature acquiring unit 10 can acquire the temperature measurement value collected by the temperature sensing probe.
  • the first determining unit 20 is configured to: after detecting that the temperature prediction has been started, and not satisfying the preset stop start condition, determine whether there is a stable flag; if yes, notify the first setting unit 30; if not, notify the first An extraction unit 40.
  • the first setting unit 30 is configured to set a temperature prediction value of the current sampling time as a temperature prediction value of the previous sampling time.
  • the first extracting unit 40 is configured to acquire temperature measurement values of the first n-1 sampling moments of the current sampling time and the current sampling time, and extract a maximum value and a minimum value of the n temperature measurement values; wherein, n Is an integer greater than 2.
  • the second determining unit 50 is configured to determine whether a difference between the maximum value and the minimum value of the n temperature measurement values is greater than a preset threshold; if yes, notify the second setting unit 60, and if not, notify the third setting Unit 70.
  • the second setting unit 60 is configured to generate a temperature prediction value of the current sampling time according to the temperature measurement value of the current sampling time and the pre-stored first offset amount.
  • the third setting unit 70 is configured to set the temperature prediction value of the current sampling time as the temperature prediction value of the previous sampling time, and generate the stable flag.
  • the output unit 80 is configured to output a temperature prediction value at the current sampling time.
  • the electronic thermometer 100 provided by the embodiment of the present invention only needs to add a suitable first offset amount to obtain a temperature prediction value when performing temperature prediction, so the required calculation amount is very Less, faster calculations, and a lot of computing resources.
  • the principle of prediction is only to add a suitable first offset to the temperature measurement, this limits the upper limit of the error to the first offset in principle, and the prediction error is limited, and it does not occur because A situation in which a large error occurs due to fitting.
  • the electronic thermometer 100 includes a third determining unit 90 and a starting unit 91 in addition to all the units in the above embodiments, wherein:
  • the third determining unit 90 is configured to: according to the temperature measurement value of the current sampling time and the temperature measurement value of at least the previous sampling time, when determining the unstarted temperature prediction, and determine whether the preset starting condition is met; if yes, The activation unit 91 is notified, and if not, the output unit 80 is notified.
  • the starting unit 91 is configured to start temperature prediction.
  • the output unit 80 is further configured to output a temperature measurement value measured at a current sampling time.
  • the temperature prediction is started: (1) the temperature measurement value is higher than 30 degrees; (2) the first derivative of the temperature measurement value Below 0.07; (3) The second derivative of the temperature measurement is less than zero.
  • the temperature measurement value of the current sampling time is x i
  • the temperature measurement value of the previous sampling time of the current sampling time is x i-1
  • the temperature measurement value of the first two sampling moments is x i-2
  • the current sampling First derivative of time x i ' x i -x i-1
  • the third determining unit 90 detects that the electronic thermometer 100 is in a relatively fast temperature rising process (eg, the first derivative of the temperature measurement value for five consecutive sampling times is higher than 0.3, or the temperature of the current sampling time)
  • the first derivative of the measured value is higher than 0.7
  • the preferred embodiment provides conditions for starting temperature prediction.
  • the electronic thermometer 100 includes, in addition to all the units of the foregoing embodiments, the following:
  • the output unit 80 is further configured to output a temperature measurement value measured at a current sampling time
  • the stopping start condition is that detecting that the continuous start time of the temperature prediction reaches a preset first time threshold or detecting that the continuous start time of the temperature prediction reaches a preset second time threshold, and the current sampling time And the difference between the maximum value and the minimum value of the temperature measurement values of the first t-1 sampling moments of the current sampling time is less than a preset threshold.
  • the preferred embodiment provides conditions for stopping the startup temperature prediction.
  • the electronic thermometer 100 includes, in addition to all the units of the foregoing embodiments, the following:
  • the second extracting unit 93 is configured to acquire temperature measurement values of the current sampling time and the first m-1 sampling moments of the current sampling time, and extract maximum and minimum values of the m temperature measurement values, where m is greater than 2 Integer
  • a correction unit 94 configured to: when the difference between the maximum value and the minimum value of the m temperature measurement values is greater than the threshold value, correct the temperature prediction value of the current sampling time to the temperature measurement value of the current sampling time Plus the second offset of the preset A quantity; wherein the second offset is less than or equal to the first offset.
  • the function is that, in view of external disturbances such as human motion during the actual measurement process, the electronic thermometer 100 may think that the temperature probe has reached thermal equilibrium or is about to reach stability in advance, so the stable mark is generated when the heat balance is not actually reached.
  • the second extraction unit 93 sets a condition for restoring the prediction, that is, the predicted temperature remains unchanged, and when the magnitude of the change of the temperature measurement within a certain time range is greater than the threshold (ie, the The difference between the maximum value and the minimum value of the m temperature measurement values is greater than the threshold value, such as the variation range of the temperature measurement value exceeds 0.05 degrees in 10 seconds, and the correction unit 94 corrects the temperature prediction value and re-predicts The predicted value of the temperature at the current sampling time.
  • the second offset at this time may be equal to the first offset or another empirical value (eg, the second offset may be 0.3).
  • the electronic thermometer 100 obtains the temperature prediction value by adding the first offset to the temperature measurement, the selection of the first offset is accurate to the prediction.
  • the speed of the degree and forecast has an important impact. The process of selecting the first offset by the embodiment of the present invention will be described in detail below.
  • the electronic thermometer 100 includes, in addition to all the units of the foregoing embodiments, the following:
  • a calculating unit 95 configured to calculate a first offset
  • the computing unit 95 specifically includes:
  • the initial condition includes a target temperature of the measured object
  • a slope calculation module 952 configured to calculate a slope of the temperature curve within a predetermined time window according to the n and the threshold;
  • a temperature value calculation module 953 configured to derive the temperature curve and obtain a predetermined x value according to the slope, and obtain a y value corresponding to the x value according to the x value;
  • the first offset calculation module 954 is configured to calculate a first offset according to the target temperature and the y value.
  • the storage medium may be a magnetic disk, an optical disk, Read-Only Memory (ROM) or Random Access Memory (RAM).

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measuring Temperature Or Quantity Of Heat (AREA)

Abstract

一种温度预测方法及电子体温计，所述温度预测方法包括：获取当前采样时刻的温度测量值（S101）；在检测到已经启动温度预测，且未满足停止启动条件后，判断是否具有稳定标记（S102）；若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值（S103）；若没有，则获取当前采样时刻及前n-1个采样时刻的温度测量值，并提取n个温度测量值中的最大值和最小值（S104）；判断n个温度测量值中的最大值与最小值的差是否大于阈值（S105）；若是，则根据温度测量值及预存的第一偏移量生成温度预测值（S106）；若否，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成稳定标记（S107）；输出当前采样时刻的温度预测值（S108）。所述电子体温计可实现快速、准确的温度预测。

Description

一种温度预测方法及电子体温计 技术领域
本发明涉及电子体温计领域，具体是一种温度预测方法及电子体温计。
背景技术
相比传统的水银玻璃体温计，电子体温计由于具有读数方便，对人体及周围环境无害(不含水银)等优点，适合于家庭使用。但是由于电子体温计的感温探头的尺寸限制，人体运动和用户佩戴的方式的不同，往往需要较长的时间才能达到稳定状态(热平衡)，即需要较长的时间才能测量出物体的真实温度。考虑到较长的测量时间会对用户体验造成一定的影响，因此一般电子体温计还利用了预测算法来***出被测物体的真实温度，以提高电子体温计的测量速度。
现有的预测算法多采用基于函数模型拟合的方法进行温度预测，预测时首先利用一定长度的历史数据和预先给定的曲线类型(如对数曲线，双曲线等)计算函数模型的参数，然后将该函数模型在某一特定时间点时的值作为被测物体的真实温度进行输出。
一般来说，函数模型包括一级函数模型和高阶函数模型，其中，一阶的函数模型的表达能力有限，而高阶函数模型极易导致过拟合，因此在训练出函数模型后，还必须对函数模型的参数进行优化。然而目前的函数模型的优化多是基于经验参数进行，适用范围有限，无法满足使用要求。
发明内容
针对上述问题，本发明提供了一种温度预测方法及电子体温计，通过设置一个合适的第一偏移量对测量得到的温度进行预测，保证了预测的快速准确，适用范围广。
本发明提供了一种温度预测方法，包括如下步骤：
获取当前采样时刻测量得到的温度测量值；
在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记；
若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值；
若没有，则获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数；
判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；
若是，则根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值；
若否，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记；
输出当前采样时刻的温度预测值。
优选地，还包括：
在检测到未启动温度预测时，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，判断是否达到预设的启动条件；
若是，则启动温度预测；
若否，则输出当前采样时刻测量得到的温度测量值。
优选地，还包括：
在满足预设的停止启动条件后，停止启动温度预测，并输出当前采样时刻测量得到的温度测量值；
其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值。
优选地，在所述若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值之后，还包括：
获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度测量值中的最大值和最小值，其中，m为大于2的整数；
当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移量；其中，所述第二偏移量小于或等于所述第一偏移量。
优选地，还包括：
计算第一偏移量；
其中，所述计算第一偏移量，具体包括：
根据初始条件求解出预设的温度曲线的曲线参数，其中，所述温度曲线为y＝a*ln(x)+b，y表示温度值，x表示时间，a，b为曲线参数，所述初始条件包括被测对象的目标温度；
根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率y’；
对所述温度曲线进行求导获得y’＝a/x，根据所述斜率y’和所述参数a获得一个x值，再 根据所述x值与所述温度曲线获得与所述x值对应的y值；
根据所述目标温度及所述y值计算得到第一偏移量。
本发明还提供了一种电子体温计，包括：
温度获取单元，用于获取当前采样时刻测量得到的温度测量值；
第一判断单元，用于在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记；若有，则通知第一设置单元；若没有，则通知第一提取单元；
所述第一设置单元，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值；
所述第一提取单元，用于获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数；
第二判断单元，用于判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；若是，则通知第二设置单元，若否，则通知第三设置单元；
第二设置单元，用于根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值；
第三设置单元，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记；
输出单元，用于输出当前采样时刻的温度预测值。
优选地，所述电子体温计还包括第三判断单元及启动单元，其中：
所述第三判断单元，用于在检测到未启动温度预测时，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，并判断是否满足预设的启动条件；若是，则通知所述启动单元，若否，则通知所述输出单元；
所述启动单元，用于启动温度预测；
所述输出单元，还用于输出当前采样时刻测量得到的温度测量值。
优选地，所述电子体温计还包括：
停止启动单元，用于在满足预设的停止启动条件后，停止启动温度预测，并通知所述输出单元；
所述输出单元，还用于输出当前采样时刻测量得到的温度测量值；
其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值。
优选地，所述电子体温计还包括：
第二提取单元，用于获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度测量值中的最大值和最小值，其中，m为大于2的整数；
修正单元，用于当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移量；其中，所述第二偏移量小于或等于所述第一偏移量。
优选地，还包括：
计算单元，用于计算第一偏移量；
其中，所述计算单元具体包括：
参数求解模块，用于根据初始条件计算得到预设的温度曲线的参数，其中，所述温度曲线为y＝a*ln(x)+b，y为温度值，x为时间，a，b为参数，所述初始条件包括被测对象的目标温度；
斜率计算模块，用于根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率；
温度值计算模块，用于对所述温度曲线进行求导并根据所述斜率获得预定的x值，再根据所述x值获得与该x值对应的y值；
第一偏移量计算模块，用于根据所述目标温度及所述y值计算得到第一偏移量。
本发明实施例提供的温度预测方法及电子体温计，在进行温度预测时，只需将温度测量值加上合适的第一偏移量，因此所需的计算量很少，计算速度快，节省了大量的计算资源，此外，而且由于预测的原理仅是将温度测量值加上合适的第一偏移量，这就从原理上限制了误差的上限为所述第一偏移量，预测误差有限，不会出现因为过拟合而导致出现很大的误差的情况。
附图说明
为了更清楚地说明本发明的技术方案，下面将对实施方式中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的温度预测方法的流程示意图。
图2是本发明实施例提供的温度测量值曲线与温度预测值曲线的对比示意图。
图3是本发明另一实施例提供的温度预测方法的流程示意图。
图4是本发明实施例提供的电子体温计的结构示意图。
图5是本发明另一实施例提供的电子体温计的结构示意图。
图6是本发明另一实施例提供的电子体温计的结构示意图。
图7是本发明另一实施例提供的电子体温计的结构示意图。
图8是本发明另一实施例提供的电子体温计的结构示意图。
图9是图8所示的计算单元的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参阅图1，本发明实施例提供一种温度预测方法，其可由电子体温计来执行，用于在使用电子体温计进行温度测量时，对被测对象(如人体)的温度进行预测，其包括如下步骤：
S101，获取当前采样时刻测量得到的温度测量值。
在本发明实施例中，所述电子体温计可通过感温探头来采集当前采样时刻的温度测量值，其中，所述感温探头可为热电阻探头，热电偶探头等，本发明不做具体限定。
在本发明实施例中，所述感温探头每隔预定时间采集一个温度测量值，例如，每秒采集一个温度测量值，当然，具体的采集频率可根据实际的需要进行设置，本发明不做具体限定。
S102，在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记。
在本发明实施例中，为了比较快速的获得被测对象的真实温度，所述电子体温计在检测的预定阶段需要启动温度预测(如在一个快速升温阶段至达到热平衡之间的时间段)，以根据历史的温度测量值来生成更接近被测对象的真实温度的温度预测值，而一旦感温探头达到热平衡后，就无需再进行预测，即在达到热平衡后，可以停止启动温度预测，直接输出感温探头测量得到的温度测量值。下面将具体描述所述电子体温计在温度预测阶段的工作过程(即在启动温度预测而尚未达到热平衡这个阶段的工作过程)。
在本发明实施例中，当所述电子体温计在检测到已经启动温度预测，且未满足预设的停止启动条件后，先判断内存中是否具有一个稳定标记。
S103，若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值。
S104，若没有，则获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数。
例如，假设所述感温探头每秒采集一个温度测量值，n为20，则所述电子体温计可获取当前采样时刻及当前采样时刻的前19个采样时刻的温度测量值，并提取这20个温度测量值 中的最大值和最小值。
S105，判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值。
在本发明实施例中，所述阈值可设置为0.05，应当理解的是，在本发明的其他实施例中，还可根据实际的需要设置所述阈值，本发明不做具体限定。
S106，若是，则根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值。
在本发明实施例中，若所述n个温度测量值中的最大值与最小值的差大于所述阈值，则说明感温探头测量的温度测量值尚有较大的波动，即还未进入较稳定的阶段，因而需要根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值，其中，所述温度预测值＝温度测量值+第一偏移量。
S107，若否，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记。
在本发明实施例中，若所述n个温度测量值中的最大值与最小值的差小于阈值时，可以认为感温探头测量的温度测量值即将达到稳定(即即将达到热平衡)，因此所述电子体温计可保持所述温度预测值不变，并生成一个稳定标记，用于标记所述感温探头即将达到稳定。此后，每个采样时刻的温度预测值均设置为上一个采样时刻的温度预测值。
在本发明实施例中，上述步骤S101-S107可用公式(1)进行表示：
Figure PCTCN2016113112-appb-000001
其中，xi为当前采样时刻的温度测量值，max(x)为n个采样时刻的温度测量值中的最大值，min(x)为n个采样时刻的温度测量值中的最小值，yi为当前时刻的温度预测值，o1为第一偏移量，θ为预设的阈值，flag为稳定标记。
S108，输出当前采样时刻的温度预测值。
综上所述，本发明实施例提供的温度预测方法，在进行温度预测时，只需将温度测量值加上合适的第一偏移量即可得到温度预测值，因此所需的计算量很少，计算速度快，节省了大量的计算资源，此外，由于温度预测的原理仅是将温度测量值加上合适的第一偏移量，这就从原理上限制了误差的上限为所述第一偏移量，预测误差有限，不会出现因为过拟合而导致出现很大的误差的情况。
需要说明的是，在一个优选实施例中，在步骤S103之后，步骤S108之前，还包括：
S109，获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度测量值中的最大值和最小值，其中，m为大于2的整数。
S110，当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移量；其中，所述第二偏移量小于或等于所述第一偏移量。
其作用在于，考虑到在实际测量过程中存在人体运动等外界干扰的因素，电子体温计可能会提前认为感温探头已经即将到达稳定，因此在未真正达到稳定时就提前生成所述稳定标记。为了避免这种情况，所述电子体温计还设置了一个恢复进入预测的条件，即预测温度仍然保持不变，但是如果一定时间范围内的温度测量值的变化幅度大于所述阈值时(即m个温度测量值中的最大值与最小值的差大于所述阈值，如10s内采集得到的温度测量值的变化幅度超过0.05度)，重新预测当前采样时刻的温度预测值。其中，所述第二偏移量可以等于所述第一偏移量，也可以为另一个经验值(如所述第二偏移量可为0.3)。
其中，本优选实施例可利用下述公式(2)进行表示：
Figure PCTCN2016113112-appb-000002
其中，o2为第二偏移量。
需要说明的是，为了避免突然加上第一偏移量而带来的温度突变，在启动温度预测和停止启动温度预测时还对预测温度曲线进行了平滑，具体地，可通过逐渐增加第一偏移量的值，实现平滑的效果，其中，所述第一偏移量可通过线性的方式进行逐渐增加，本发明不做具体限定。
如图2所示，在刚启动温度预测时，所述第一偏移量逐渐增加至0.7度，而非突变。在第10分钟左右时尽管先给出了一个错误的稳定温度，但是在运行一小段时间后恢复了计算偏移量(此时第二偏移量的值为0.3度)，并且最终给出了正确的稳定温度。本发明实施例中，通过设置了恢复预测条件对温度预测值进行修正，使得本发明的温度预测具有较强的抗干扰能力，在有干扰的情况下，温度预测值随着温度测量值的变化而变化，当干扰减少或消失，温度测量值的波动较小时，温度预测值将重新达到稳定。
需要说明的是，在一个优选实施例中，由于电子体温计是通过温度测量值加上所述第一偏移量来获得温度预测值，因而所述第一偏移量的选取对预测的准确度及预测的速度有重要影响。下面将详细描述本发明实施例选取所述第一偏移量的过程。具体地：
在一个优选实施例中，所述的温度预测方法还包括：
S111，计算第一偏移量；
其中，具体包括：
S1111，根据初始条件求解出预设的温度曲线的参数，其中，所述温度曲线为y＝a*ln(x)+b，y表示温度值，x表示时间，a，b为参数，所述初始条件包括被测对象的目标温度。
S1112，根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率y’。
S1113，对所述温度曲线进行求导获得y’＝a/x，根据所述斜率y’和所述参数a获得一个x值，再根据所述x值与所述温度曲线获得与所述x值对应的y值。
S1114，根据所述目标温度及所述y值计算得到第一偏移量。
例如：对于体温曲线
y＝a*ln(x)+b          (3)
x为时间(s)，y为温度，a，b为曲线参数。因为通常情况下电子体温计在15min(900s)左右到达稳定，即5min内升温不超过0.15度(1200s时体温不超过37.15度)。设目标温度为人体的正常体温37度，因此可以建立方程组
y1＝a*ln(x1)+b
y2＝a*ln(x2)+b           (4)
将初始条件：y1＝37，x1＝900；y2＝37.15，x2＝1200代入方程组(3)，求得a＝0.5214，b＝33.4532。
在本发明实施例中，虽然电子体温计的升温曲线为对数曲线，但是在一个很短的时间窗口内，其升温曲线可以由直线近似表达。如果选择20s(Δt，即n个采样时刻所需的时间)内升温幅度0.05度(Δy，即所述阈值)为预测温度是否启动的判断条件，其该时间窗口内的斜率为：
y′＝Δy/Δt＝0.05/20＝0.0025         (5)
由于对数曲线的一阶导数公式为
y′＝a/x                      (6)
将y′＝0.0025，a＝0.5214代入公式(6)，可求得x＝208.56。根据公式(3)，此时的测量温度为y＝a*ln(x)+b＝0.5214*ln(208.56)+33.4532＝36.24，与目标温度37度相差0.76度，即第一偏移量为0.76度。
考虑到理论曲线与实际测量时的误差，可以将所述第一偏移量稍稍缩小到0.7度。
当然，为了提高预测的速度，可以选择一个较大的第一偏移量，此时判断条件中的时间窗口(Δt)和阈值(Δy)也应当做出相应的改变。
请一并参阅图3，图3是本发明另一实施例提供的温度预测方法的流程示意图。其至少包括如下步骤：
S201，获取当前采样时刻测量得到的温度测量值。
S202，检测是否已经启动温度预测。若是，执行S203，若否，执行S204。
S203，判断是否满足预设的停止启动条件，若满足，执行S205，若不满足，执行S206。
其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值(如温度预测的启动时间超过1小时)或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值，如温度预测启动超过20分钟，且一定时间的温度测量值的变化幅度小于阈值(例如10秒内温度测量值的变化幅度小于0.05度)。
S204，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，判断是否达到预设的启动条件；若是，则跳转至步骤S208；若否，则执行S205。
例如，若当前采样时刻的温度测量值同时满足如下启动条件时启动温度预测。(1)温度测量值高于30度；(2)温度测量值的一阶导数低于0.07；(3)温度测量值二阶导数小于0。其中，设当前采样时刻的温度测量值为xi，当前采样时刻的前一个采样时刻的温度测量值为xi-1，前两个采样时刻的温度测量值为xi-2，则当前采样时刻的一阶导数xi’＝xi-xi-1，二阶导数xi”＝(xi-xi-1)-(xi-1-xi-2)＝(xi+xi-2)-2*xi-1。
又或者，当检测到一个较快的升温过程(如连续5个采样时刻的温度测量值的一阶导数均高于0.3，或者当前采样时刻的温度测量值的一阶导数高于0.7)时，判定达到预设的启动条件，并启动温度预测。
S205，输出当前采样时刻测量得到的温度测量值。
S206，判断是否具有稳定标记；若有，执行S207，若没有，执行S208。
S207，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值；
S208，则获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数；
S209，判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；若否，则执行S210，若是，则执行S211。
S210，则根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值。
S211，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记。
S212，输出当前采样时刻的温度预测值。
本发明实施例提供的温度预测方法，提供了进入温度预测的启动条件和温度预测的停止启动条件，且在温度预测时，只需将温度测量值加上合适的第一偏移量即可生成温度预测值，因此所需的计算量很少，计算速度快，节省了大量的计算资源。此外，由于预测的原理仅是将温度测量值加上合适的第一偏移量，这就从原理上限制了误差的上限为所述第一偏移量，预测误差有限，不会出现因为过拟合而导致出现很大的误差的情况。
请参阅图4，图4是本发明实施例提供的电子体温计的结构示意图。所述电子体温计100可用于执行上述的温度预测方法，其包括：
温度获取单元10，用于获取当前采样时刻测量得到的温度测量值。
其中，所述温度获取单元10可获取感温探头采集到的温度测量值。
第一判断单元20，用于在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记；若有，则通知第一设置单元30；若没有，则通知第一提取单元40。
所述第一设置单元30，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值。
所述第一提取单元40，用于获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数。
第二判断单元50，用于判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；若是，则通知第二设置单元60，若否，则通知第三设置单元70。
第二设置单元60，用于根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值。
第三设置单元70，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记。
输出单元80，用于输出当前采样时刻的温度预测值。
综上所述，本发明实施例提供的电子体温计100，在进行温度预测时，只需将温度测量值加上合适的第一偏移量即可获得温度预测值，因此所需的计算量很少，计算速度快，节省了大量的计算资源。此外，由于预测的原理仅是将温度测量值加上合适的第一偏移量，这就从原理上限制了误差的上限为所述第一偏移量，预测误差有限，不会出现因为过拟合而导致出现很大的误差的情况。
请一并参阅图5，在一个优选实施例中，所述电子体温计100除了包括上述实施例的所有单元外，还包括第三判断单元90及启动单元91，其中：
所述第三判断单元90，用于在检测到未启动温度预测时，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，并判断是否满足预设的启动条件；若是，则通知所述启动单元91，若否，则通知所述输出单元80。
所述启动单元91，用于启动温度预测。
所述输出单元80，还用于输出当前采样时刻测量得到的温度测量值。
其中，若所述第三判断单元90在判断当前采样时刻的温度测量值同时满足如下启动条件时启动温度预测：(1)温度测量值高于30度；(2)温度测量值的一阶导数低于0.07；(3)温度测量值二阶导数小于0。其中，设当前采样时刻的温度测量值为xi，当前采样时刻的前一个采样时刻的温度测量值为xi-1，前两个采样时刻的温度测量值为xi-2，则当前采样时刻的一阶导数xi’＝xi-xi-1，二阶导数xi”＝(xi-xi-1)-(xi-1-xi-2)＝(xi+xi-2)-2*xi-1。
又或者，当所述第三判断单元90在检测到电子体温计100处于一个较快的升温过程(如连续5个采样时刻的温度测量值的一阶导数均高于0.3，或者当前采样时刻的温度测量值的一阶导数高于0.7)时，判定达到预设的启动条件，并启动温度预测。
本优选实施例提供了启动温度预测的条件。
请一并参阅图6，在一个优选实施例中，所述电子体温计100除了包括上述实施例的所有单元外，还包括：
停止启动单元92，用于在满足预设的停止启动条件后，停止启动温度预测，并通知所述输出单元80；
所述输出单元80，还用于输出当前采样时刻测量得到的温度测量值；
其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值。
本优选实施例提供了停止启动温度预测的条件。
请一并参阅图7，在一个优选实施例中，所述电子体温计100除了包括上述实施例的所有单元外，还包括：
第二提取单元93，用于获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度测量值中的最大值和最小值，其中，m为大于2的整数；
修正单元94，用于当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移 量；其中，所述第二偏移量小于或等于所述第一偏移量。
其作用在于，考虑到在实际测量过程中存在人体运动等外界干扰，电子体温计100可能会提前认为感温探头已经达到热平衡或即将到达稳定，因此在未真正达到热平衡时就生成所述稳定标记，为了防止这种情况，所述第二提取单元93设置了一个恢复进入预测的条件，即预测温度保持不变，且一定时间范围内的温度测量值的变化幅度大于所述阈值时(即所述m个温度测量值中的最大值与最小值的差大于所述阈值，如10s内温度测量值的变化幅度超过0.05度)，由所述修正单元94对所述温度预测值进行修正，重新预测当前采样时刻的温度预测值。此时的第二偏移量可以等于所述第一偏移量，也可以为另一个经验值(如所述第二偏移量可为0.3)。
需要说明的是，在一个优选实施例中，由于电子体温计100是通过温度测量值加上所述第一偏移量来获得温度预测值，因而所述第一偏移量的选取对预测的准确度及预测的速度有重要影响。下面将详细描述本发明实施例选取所述第一偏移量的过程。
请一并参阅图8及图9，在一个优选实施例中，所述电子体温计100除了包括上述实施例的所有单元外，还包括：
计算单元95，用于计算第一偏移量；
其中，所述计算单元95具体包括：
参数求解模块951，用于根据初始条件计算得到预设的温度曲线的参数，其中，所述温度曲线为y＝a*ln(x)+b，y为温度值，x为时间，a，b为参数，所述初始条件包括被测对象的目标温度；
斜率计算模块952，用于根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率；
温度值计算模块953，用于对所述温度曲线进行求导并根据所述斜率获得预定的x值，再根据所述x值获得与该x值对应的y值；
第一偏移量计算模块954，用于根据所述目标温度及所述y值计算得到第一偏移量。
以上所揭露的仅为本发明一种较佳实施例而已，当然不能以此来限定本发明之权利范围，本领域普通技术人员可以理解实现上述实施例的全部或部分流程，并依本发明权利要求所作的等同变化，仍属于发明所涵盖的范围。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述的存储介质可为磁碟、光盘、 只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。

Claims (10)

  1. 一种温度预测方法，其特征在于，包括如下步骤：
    获取当前采样时刻测量得到的温度测量值；
    在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记；
    若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值；
    若没有，则获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数；
    判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；
    若是，则根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值；
    若否，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记；
    输出当前采样时刻的温度预测值。
  2. 根据权利要求1所述的温度预测方法，其特征在于，还包括：
    在检测到未启动温度预测时，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，判断是否达到预设的启动条件；
    若是，则启动温度预测；
    若否，则输出当前采样时刻测量得到的温度测量值。
  3. 根据权利要求1所述的温度预测方法，其特征在于，还包括：
    在满足预设的停止启动条件后，停止启动温度预测，并输出当前采样时刻测量得到的温度测量值；
    其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值。
  4. 根据权利要求1所述的温度预测方法，其特征在于，在所述若有，则将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值之后，还包括：
    获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度 测量值中的最大值和最小值，其中，m为大于2的整数；
    当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移量；其中，所述第二偏移量小于或等于所述第一偏移量。
  5. 根据权利要求1-4任意一项所述的温度预测方法，其特征在于，还包括：
    计算第一偏移量；
    其中，所述计算第一偏移量，具体包括：
    根据初始条件求解出预设的温度曲线的曲线参数，其中，所述温度曲线为y＝a*ln(x)+b，y表示温度值，x表示时间，a，b为曲线参数，所述初始条件包括被测对象的目标温度；
    根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率y’；
    对所述温度曲线进行求导获得y’＝a/x，根据所述斜率y’和所述参数a获得一个x值，再根据所述x值与所述温度曲线获得与所述x值对应的y值；
    根据所述目标温度及所述y值计算得到第一偏移量。
  6. 一种电子体温计，其特征在于，包括：
    温度获取单元，用于获取当前采样时刻测量得到的温度测量值；
    第一判断单元，用于在检测到已经启动温度预测，且未满足预设的停止启动条件后，判断是否具有稳定标记；若有，则通知第一设置单元；若没有，则通知第一提取单元；
    所述第一设置单元，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值；
    所述第一提取单元，用于获取当前采样时刻及当前采样时刻的前n-1个采样时刻的温度测量值，并提取这n个温度测量值中的最大值和最小值；其中，n为大于2的整数；
    第二判断单元，用于判断所述n个温度测量值中的最大值与最小值的差是否大于预设的阈值；若是，则通知第二设置单元，若否，则通知第三设置单元；
    第二设置单元，用于根据当前采样时刻的温度测量值及预存的第一偏移量生成当前采样时刻的温度预测值；
    第三设置单元，用于将当前采样时刻的温度预测值设置为上一个采样时刻的温度预测值，并生成所述稳定标记；
    输出单元，用于输出当前采样时刻的温度预测值。
  7. 根据权利要求6所述的电子体温计，其特征在于，所述电子体温计还包括第三判断单元及启动单元，其中：
    所述第三判断单元，用于在检测到未启动温度预测时，根据当前采样时刻的温度测量值及至少前一个采样时刻的温度测量值，并判断是否满足预设的启动条件；若是，则通知所述启动单元，若否，则通知所述输出单元；
    所述启动单元，用于启动温度预测；
    所述输出单元，还用于输出当前采样时刻测量得到的温度测量值。
  8. 根据权利要求6所述的电子体温计，其特征在于，所述电子体温计还包括：
    停止启动单元，用于在满足预设的停止启动条件后，停止启动温度预测，并通知所述输出单元；
    所述输出单元，还用于输出当前采样时刻测量得到的温度测量值；
    其中，所述停止启动条件为检测到所述温度预测的持续启动时间达到预设的第一时间阈值或者检测到所述温度预测的持续启动时间达到预设的第二时间阈值，且当前采样时刻及当前采样时刻的前t-1个采样时刻的温度测量值的最大值与最小值的差小于预设的阈值。
  9. 根据权利要求6所述的电子体温计，其特征在于，所述电子体温计还包括：
    第二提取单元，用于获取当前采样时刻及当前采样时刻的前m-1个采样时刻的温度测量值，提取这m个温度测量值中的最大值和最小值，其中，m为大于2的整数；
    修正单元，用于当所述m个温度测量值中的最大值与最小值的差大于所述阈值时，将所述当前采样时刻的温度预测值修正为所述当前采样时刻的温度测量值加上预设的第二偏移量；其中，所述第二偏移量小于或等于所述第一偏移量。
  10. 根据权利要求6至9任意一项所述的电子体温计，其特征在于，还包括：
    计算单元，用于计算第一偏移量；
    其中，所述计算单元具体包括：
    参数求解模块，用于根据初始条件计算得到预设的温度曲线的参数，其中，所述温度曲线为y＝a*ln(x)+b，y为温度值，x为时间，a，b为参数，所述初始条件包括被测对象的目标温度；
    斜率计算模块，用于根据所述n及所述阈值计算所述温度曲线在预定时间窗口内的斜率；
    温度值计算模块，用于对所述温度曲线进行求导并根据所述斜率获得预定的x值，再根 据所述x值获得与该x值对应的y值；
    第一偏移量计算模块，用于根据所述目标温度及所述y值计算得到第一偏移量。
PCT/CN2016/113112 2016-04-29 2016-12-29 一种温度预测方法及电子体温计 WO2017185796A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610286218.7A CN105956399B (zh) 2016-04-29 2016-04-29 一种温度预测方法及电子体温计
CN201610286218.7 2016-04-29

Publications (1)

Publication Number Publication Date
WO2017185796A1 true WO2017185796A1 (zh) 2017-11-02

Family

ID=56914586

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/113112 WO2017185796A1 (zh) 2016-04-29 2016-12-29 一种温度预测方法及电子体温计

Country Status (2)

Country Link
CN (1) CN105956399B (zh)
WO (1) WO2017185796A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113970382A (zh) * 2020-07-06 2022-01-25 浙江宇视科技有限公司 一种温度检测方法、装置、介质及电子设备
CN115219062A (zh) * 2022-04-18 2022-10-21 广州汽车集团股份有限公司 一种设备的温度确定方法、装置、设备及存储介质
CN116086632A (zh) * 2023-01-28 2023-05-09 天翼云科技有限公司 数据处理方法、装置、电子设备及介质
CN117516629A (zh) * 2023-11-13 2024-02-06 美特文博（天津）科技有限公司 一种文物存储环境检测***和储藏及展示装备

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105956399B (zh) * 2016-04-29 2018-02-09 广州视源电子科技股份有限公司 一种温度预测方法及电子体温计
CN106473708B (zh) * 2016-11-29 2019-03-29 广州视源电子科技股份有限公司 体温预测方法、装置及电子体温计
CN106510651B (zh) * 2016-12-09 2019-04-16 广州视源电子科技股份有限公司 一种体温预测算法的评估方法及装置
CN108345953B (zh) * 2017-01-24 2021-12-17 上海温尔信息科技有限公司 温度预测方法及装置
CN108871615B (zh) * 2018-07-24 2020-06-05 杭州质子科技有限公司 一种基于长时间监测的人体腋下体温抗干扰处理方法
CN109813264A (zh) * 2019-02-21 2019-05-28 重庆潍柴发动机有限公司 测量结果误差评估的方法及装置
CN110434154A (zh) * 2019-08-15 2019-11-12 中国科学院武汉岩土力学研究所 好氧通风型垃圾填埋场温度预测方法
CN113520306B (zh) * 2020-04-17 2024-03-22 青岛海尔空调器有限总公司 一种人体睡眠状态的监测方法和智能家居装置
CN111982293B (zh) * 2020-07-20 2022-07-15 北京声智科技有限公司 体温测量方法、装置、电子设备及存储介质
CN113029392B (zh) * 2021-04-15 2024-05-31 高创(苏州)电子有限公司 一种体温测量方法、装置及电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101115979A (zh) * 2005-03-29 2008-01-30 西铁城控股株式会社 电子体温计
CN101435727A (zh) * 2007-11-12 2009-05-20 深圳迈瑞生物医疗电子股份有限公司 温度预测方法及装置
CN103234647A (zh) * 2013-04-16 2013-08-07 广东欧珀移动通信有限公司 一种嵌入式***的温度校准方法及***
US20140149065A1 (en) * 2010-05-27 2014-05-29 Exergen Corporation Method and apparatus for accurate detection of fever
CN105371968A (zh) * 2015-11-20 2016-03-02 广州视源电子科技股份有限公司 电子温度计控制方法及其装置
CN105956399A (zh) * 2016-04-29 2016-09-21 广州视源电子科技股份有限公司 一种温度预测方法及电子体温计

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101115979A (zh) * 2005-03-29 2008-01-30 西铁城控股株式会社 电子体温计
CN101435727A (zh) * 2007-11-12 2009-05-20 深圳迈瑞生物医疗电子股份有限公司 温度预测方法及装置
US20140149065A1 (en) * 2010-05-27 2014-05-29 Exergen Corporation Method and apparatus for accurate detection of fever
CN103234647A (zh) * 2013-04-16 2013-08-07 广东欧珀移动通信有限公司 一种嵌入式***的温度校准方法及***
CN105371968A (zh) * 2015-11-20 2016-03-02 广州视源电子科技股份有限公司 电子温度计控制方法及其装置
CN105956399A (zh) * 2016-04-29 2016-09-21 广州视源电子科技股份有限公司 一种温度预测方法及电子体温计

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113970382A (zh) * 2020-07-06 2022-01-25 浙江宇视科技有限公司 一种温度检测方法、装置、介质及电子设备
CN115219062A (zh) * 2022-04-18 2022-10-21 广州汽车集团股份有限公司 一种设备的温度确定方法、装置、设备及存储介质
CN116086632A (zh) * 2023-01-28 2023-05-09 天翼云科技有限公司 数据处理方法、装置、电子设备及介质
CN116086632B (zh) * 2023-01-28 2023-08-04 天翼云科技有限公司 数据处理方法、装置、电子设备及介质
CN117516629A (zh) * 2023-11-13 2024-02-06 美特文博（天津）科技有限公司 一种文物存储环境检测***和储藏及展示装备
CN117516629B (zh) * 2023-11-13 2024-04-26 美特文博（天津）科技有限公司 一种文物存储环境检测***和储藏及展示装备

Also Published As

Publication number Publication date
CN105956399B (zh) 2018-02-09
CN105956399A (zh) 2016-09-21

Similar Documents

Publication Publication Date Title
WO2017185796A1 (zh) 一种温度预测方法及电子体温计
WO2017113875A1 (zh) 预测温度的方法及其***
CN108268082B (zh) 功率控制方法及装置
JP6660613B2 (ja) 断熱性能推定装置、及び、断熱性能推定方法
RU2007140372A (ru) Устройство и способ прогнозирования температуры человека
US10261136B2 (en) Battery degradation degree estimation device and battery degradation degree estimation method
CN105403325B (zh) 预测移动设备附近的环境温度的方法、计算机程序产品和移动设备
CN104729748B (zh) 用于确定移动设备的环境温度的方法
US20140316721A1 (en) Gas flow detecting device and gas flow detecting method
WO2017211071A1 (zh) 预测温度的方法及其装置
KR102046697B1 (ko) 환경 센서 및 환경 파라미터의 측정과 예측 방법
CN110462538B (zh) 信息处理装置、信息处理方法以及存储介质
CN105286812B (zh) 一种体温测量方法和装置
US20190025104A1 (en) Thermal flowmeter and flow rate correction method
CN105342579B (zh) 一种体温测量方法和装置
CN113970382B (zh) 一种温度检测方法、装置、介质及电子设备
CN104598361B (zh) 一种性能监控方法和装置
CN107729625A (zh) 一种对设备运行发热造成的温度测量误差进行补偿的方法及装置
JP2006010677A (ja) 情報処理装置の温度センサー装置
JP2010122163A (ja) 電子体温計
JP6385777B2 (ja) 電子体温計
JP6278402B2 (ja) 電子体温計
JP6269678B2 (ja) ファジイ制御装置及びファジイ制御方法
JP5235963B2 (ja) 温度測定装置及びこの温度測定装置を用いた空気調和機
JP7473074B2 (ja) 温度推定方法、温度推定プログラムおよび温度推定装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16900300

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 16900300

Country of ref document: EP

Kind code of ref document: A1