Method for determining temporal physiological characteristics and thermal comfort prediction under target environment

Abstract

The present invention provides a method for determining temporal physiological characteristics in a target environment and a method for predicting thermal comfort, which can be applied to the field of thermal comfort prediction technology. The method comprises: sorting I characteristic indicators according to the similarity between temporal physiological characteristic data of characteristic indicators of a sampling object in a target environment and a thermal comfort temporal label to obtain a sorting result; combining multiple temporal physiological characteristic data based on a preset combination rule and the sorting result to obtain I temporal physiological characteristic sets; determining the i-th indicator performance change between the i-th temporal physiological characteristic set and the i-1-th temporal physiological characteristic set; when the i-th indicator performance change is less than the i-1-th indicator performance change, determining a target physiological characteristic set from the first to i-th temporal physiological characteristic sets; and establishing a mapping relationship between the group type of the sampling object and the target physiological characteristic set to obtain an object feature mapping relationship table.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal comfort prediction, and in particular to a method for determining temporal physiological characteristics and a method for predicting thermal comfort in a target environment. Background Art

[0002] Thermal comfort, a core element connecting human needs with energy efficiency, is crucial for optimizing thermal environment regulation, improving energy management efficiency, and safeguarding human health. Human thermal comfort is not an immediate response to the thermal environment or metabolic levels, but rather a multi-layered process of thermal regulation, physiological feedback, and sensory integration. It is also influenced by past thermal experiences, resulting in significant time lags and cumulative characteristics.

[0003] However, existing modeling methods that aim to discretely predict thermal comfort states at a single point in time are unable to effectively describe the continuous evolution of thermal comfort states and the potential state transition process. In addition, a large amount of multi-source heterogeneous information, such as multi-dimensional environmental parameters, physiological signals, and subjective feedback, is used as model input, which increases the computational burden of the model and reduces the prediction accuracy. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method for determining temporal physiological characteristics and a method for predicting thermal comfort in a target environment.

[0005] According to a first aspect of the present invention, a method for determining a temporal physiological characteristic in a target environment is provided, comprising: for each of I characteristic indicators, sorting the I characteristic indicators according to the similarity between the temporal physiological characteristic data of the characteristic indicator and the thermal comfort temporal label of the sampling object in the target environment to obtain a sorting result; combining multiple temporal physiological characteristic data based on a preset combination rule and the sorting result to obtain I temporal physiological characteristic sets, where I is an integer greater than 1; determining a performance change of an i-th indicator between the i-th temporal physiological characteristic set and the i-1-th temporal physiological characteristic set, The i-th indicator performance change represents the degree of change between the accuracy of evaluating the thermal comfort of the sampling object based on the i-1th time-series physiological feature set and the accuracy of evaluating the thermal comfort of the sampling object based on the i-th time-series physiological feature set. The i-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located at the 1st position to the i-th position in the sorting result, and the i-1th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located at the 1st position to the i-1th position in the sorting result. When the i-th indicator performance change is less than the i-1th indicator performance change, the target physiological feature set is determined from the 1st to i-th time-series physiological feature sets, wherein the target physiological feature set includes the time-series physiological feature data corresponding to at least one target feature indicator. A mapping relationship between the group type of the sampling object and the target physiological feature set is established to obtain an object feature mapping relationship table.

[0006] The second aspect of the present invention provides a thermal comfort prediction method based on temporal physiological characteristics, comprising: in response to a thermal comfort prediction request for a target object in a target environment, determining the target object group type to which the target object belongs; determining the target feature index corresponding to the target object based on an object feature mapping relationship table and the target object group type, wherein the object feature mapping relationship table is a correspondence between sampling objects belonging to different object group types and target feature indicators, and the object feature mapping relationship table is obtained based on the temporal physiological feature determination method under the above-mentioned target environment; using a prediction model to process the temporal physiological feature data of the target feature index corresponding to the target object to obtain a thermal comfort prediction result for the target object.

[0007] A third aspect of the present invention provides a device for determining a temporal physiological characteristic in a target environment, comprising: a sorting module for sorting, for each of I characteristic indicators, the I characteristic indicators according to the similarity between the temporal physiological characteristic data of the characteristic indicator and the thermal comfort temporal label of the sampling object in the target environment, to obtain a sorting result; a combining module for combining multiple temporal physiological characteristic data based on a preset combination rule and the sorting result to obtain I temporal physiological characteristic sets, where I is an integer greater than 1; a first determining module for determining an i-th indicator performance change between the i-th temporal physiological characteristic set and the i-1-th temporal physiological characteristic set. The i-th indicator performance change represents the degree of change in the accuracy of evaluating the thermal comfort of the sampling object based on the i-th time-series physiological feature set compared to the i-1-th time-series physiological feature set, the i-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-th position in the sorting result, and the i-1-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-1-th position in the sorting result; the second determination module is used to determine the target physiological feature set from the i time-series physiological feature sets when the i-th indicator performance change is less than the i-1-th indicator performance change, wherein the target physiological feature set includes the time-series physiological feature data corresponding to at least one target feature indicator; the mapping module is used to establish a mapping relationship between the group type of the sampling object and the target physiological feature set to obtain an object feature mapping relationship table.

[0008] The fourth aspect of the present invention provides a thermal comfort prediction device based on temporal physiological characteristics, including: a response module, used to respond to a thermal comfort prediction request for a target object in a target environment, and determine the target object group type to which the target object belongs; a third determination module, used to determine the target feature index corresponding to the target object based on the object feature mapping relationship table and the target object group type, wherein the object feature mapping relationship table is the correspondence between sampling objects belonging to different object group types and the target feature index, and the object feature mapping relationship table is obtained according to the temporal physiological feature determination method under the above-mentioned target environment; a prediction module, used to use a prediction model to process the temporal physiological feature data of the target feature index corresponding to the target object to obtain a thermal comfort prediction result for the target object.

[0009] The fifth aspect of the present invention provides an electronic device comprising: one or more processors; a memory for storing one or more programs, wherein, when the one or more programs are executed by one or more processors, the one or more processors execute the above-mentioned method for determining temporal physiological characteristics under the target environment.

[0010] According to the method for determining time-series physiological characteristics and predicting thermal comfort in a target environment provided by the present invention, the similarity between the time-series physiological characteristic data of each characteristic indicator and the thermal comfort time-series label is calculated, and multiple time-series physiological characteristic data are combined to obtain I time-series physiological characteristic sets. The difference between the performance change of the i-th indicator and the performance change of the i-1th indicator is then compared. When the i-1th time-series physiological characteristic set is determined as the target physiological characteristic set, thermal comfort prediction for the remaining i-i time-series physiological characteristic sets is stopped. This method can achieve efficient feature selection when faced with a large number of complex time-series physiological characteristics, while solving the technical problem that the thermal comfort prediction model needs to process all target time-series physiological characteristic data, resulting in high model overhead. The construction of the object feature mapping relationship table facilitates thermal comfort prediction for objects in the same target environment based directly on a small amount of time-series physiological characteristic data corresponding to the target characteristic indicators with low redundancy. This effectively captures non-steady-state response characteristics, improves the modeling expression capability and prediction accuracy of the thermal comfort response process driven by coupled disturbances, and accurately predicts the dynamic evolution of the thermal comfort state over a continuous time period, effectively improving the adaptability and computational efficiency of the prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other objects, features and advantages of the present invention will become more apparent from the following description of the embodiments of the present invention with reference to the accompanying drawings.

[0012] Figure 1 A flow chart of a method for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown.

[0013] Figure 2 An example diagram of a prediction model according to an embodiment of the present invention is shown.

[0014] Figure 3 An example diagram showing the division of different functional spaces in a full-process simulation experiment of a high-speed railway station ride according to an embodiment of the present invention is shown.

[0015] Figure 4 An example diagram of obtaining simulation experiment data according to an embodiment of the present invention is shown.

[0016] Figure 5 A flow chart of a thermal comfort prediction method based on temporal physiological characteristics according to an embodiment of the present invention is shown.

[0017] Figure 6 A structural block diagram of a device for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown.

[0018] Figure 7 The figure shows a structural block diagram of a thermal comfort prediction device based on time series physiological characteristics according to an embodiment of the present invention.

[0019] Figure 8 A block diagram of an electronic device suitable for implementing a method for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0020] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concept of the present invention.

[0021] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise," "include," etc. used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0022] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0023] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0024] Related research has found that in the field of thermal comfort prediction technology, existing dynamic thermal comfort prediction methods based on physiological parameters generally have difficulty accurately reflecting the continuous change trends and temporal inertia characteristics of thermal comfort states, especially when dealing with non-periodic changes or sudden changes. In addition, existing methods generally lack efficient feature selection and dimensionality reduction mechanisms, resulting in high feature dimensions and a large amount of redundant information, which in turn affects modeling efficiency and weakens the model's predictive performance. Therefore, in the context of multimodal physiological characteristics, how to use highly correlated low-dimensional temporal physiological characteristics to achieve efficient and accurate prediction of dynamic thermal comfort over continuous time periods has become a technical problem that needs to be solved urgently.

[0025] In view of this, embodiments of the present invention provide a method for determining time-series physiological characteristics and a method for predicting thermal comfort in a target environment. The method includes: sorting I characteristic indicators based on the similarity between the time-series physiological characteristic data for the characteristic indicators and the thermal comfort time-series labels of the sampling subjects in the target environment to obtain a sorting result; combining multiple time-series physiological characteristic data based on a preset combination rule and the sorting result to obtain I time-series physiological characteristic sets; determining the performance change of the i-th indicator between the i-th time-series physiological characteristic set and the i-1-th time-series physiological characteristic set; if the performance change of the i-th indicator is less than the performance change of the i-1-th indicator, determining a target physiological characteristic set from the first to i-th time-series physiological characteristic sets; and establishing a mapping relationship between the group type of the sampling subjects and the target physiological characteristic set to obtain an object feature mapping relationship table.

[0026] In the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0027] In scenarios where personal information is used for automated decision-making, the methods, devices, and systems provided by embodiments of the present invention provide users with corresponding operational portals, allowing them to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The term "automated decision-making" herein refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, or credit status through computer programs and making decisions. The term "expert decision-making" herein refers to the activity of decision-making by individuals who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0028] It should be noted that the sequence numbers of the operations in the following method are only used to indicate the operation for the purpose of description, and should not be regarded as indicating the order in which the operations should be performed. Unless explicitly stated, the method does not need to be performed in the order shown.

[0029] Figure 1 A flow chart of a method for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown.

[0030] like Figure 1 As shown, the method 100 for determining temporal physiological characteristics under a target environment includes operations S110 to S150.

[0031] In operation S110, for each of the I characteristic indicators, the I characteristic indicators are sorted according to the similarity between the time-series physiological characteristic data of the characteristic indicator and the thermal comfort time-series label of the sampling object in the target environment to obtain a sorting result.

[0032] Optionally, the sampling subjects can be healthy adults. The target environment is the environment to be studied. Environmental factors of the target environment include air temperature, radiant temperature, humidity, and wind speed. Time windows or functional spaces can be divided based on the environmental factors of the target environment and the metabolic level of the sampling subjects. The metabolic level of the sampling subjects is assessed by behavioral activity patterns and intensity.

[0033] Optionally, based on different time windows or functional spaces, each time series physiological characteristic data and thermal comfort time series label of the sampling object is collected.

[0034] Optionally, the characteristic indicators represent physiological characteristics of the sampled subject, and the time series physiological characteristic data represents time series data obtained by sampling the same characteristic indicator multiple times within a time window. The characteristic indicators may include skin thermal imaging features, skin temperature features, heart rate features, heart rate variability features, electrocardiogram (ECG) features, metabolic rate features, blood pressure features, skin impedance features, skin conductivity features, blood oxygen saturation features, respiratory rate features, oxygen consumption rate features, carbon dioxide partial pressure features, sweat rate features, EEG features, sensory nerve conduction velocity features, etc.

[0035] Optionally, thermal comfort time series labels can be collected through questionnaires using a standardized thermal comfort assessment scale, and the subjective thermal comfort scores of the sampled subjects in different time windows or functional spaces can be recorded multiple times. In a standardized thermal comfort assessment scale, labels such as thermal sensation, thermal preference, thermal acceptability, and thermal satisfaction can be used as thermal comfort time series labels.

[0036] Optionally, the categories of the thermal sensation time series label may be very cold, cold, cool, slightly cool, moderate, slightly warm, warm, hot, or very hot.

[0037] Optionally, a similarity function is used to calculate the similarity between the time-series physiological characteristic data of each characteristic indicator and the thermal comfort time-series label to obtain the similarity corresponding to each characteristic indicator. The multiple characteristic indicators can be sorted in descending order according to the multiple similarities to obtain a sorting result.

[0038] Optionally, the sorting result includes multiple feature indices having a sequence relationship.

[0039] In operation S120, based on a preset combination rule and a sorting result, a plurality of time-series physiological characteristic data are combined to obtain a time-series physiological characteristic set.

[0040] Optionally, I is an integer greater than 1.

[0041] Optionally, the preset combination rule may be to add the time series physiological feature data one by one from large to small to construct I time series physiological feature set.

[0042] For example, the sorting result is (skin temperature feature, heart rate feature, heart rate variability feature). According to the preset combination rules, the time series physiological feature data under multiple feature indicators are combined to obtain three time series physiological feature sets: (skin temperature feature data), (skin temperature feature data, heart rate data), (skin temperature feature data, heart rate data, heart rate variability feature data).

[0043] In operation S130 , an i-th indicator performance change between the i-th time-series physiological feature set and the i−1-th time-series physiological feature set is determined.

[0044] Optional, The change in the performance of the i-th indicator represents the degree of change in the accuracy of evaluating the thermal comfort of the sampled object based on the i-th time-series physiological feature set compared to the i-1-th time-series physiological feature set.

[0045] Optionally, the i-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located at the 1st position to the i-th position in the sorting result, and the i-1-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located at the 1st position to the i-1-th position in the sorting result.

[0046] For example, taking the above-mentioned sorting result of (skin temperature feature, heart rate feature, heart rate variability feature) as an example, the feature indicator at the first position in the sorting result is the skin temperature feature, the feature indicator at the second position is the heart rate feature, and the feature indicator at the third position is the heart rate variability feature. The third time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators at the first position to the third position in the sorting result, namely (skin temperature feature data, heart rate data, heart rate variability feature data).

[0047] Optionally, for the i-th time-series physiological feature set, the thermal comfort prediction model can be used to process the time-series physiological feature data included in the i-th time-series physiological feature set to obtain a thermal comfort prediction result, and an accuracy evaluation result can be obtained based on the thermal comfort prediction result and the thermal comfort time-series label.

[0048] Optionally, for the i-1th time-series physiological feature set, the thermal comfort prediction model can be used to process the time-series physiological feature data included in the i-1th time-series physiological feature set to obtain a thermal comfort prediction result, and an accuracy evaluation result can be obtained based on the thermal comfort prediction result and the thermal comfort time-series label.

[0049] Optionally, the i-th indicator performance change is obtained according to the difference between the accuracy evaluation result corresponding to the i-th time-series physiological feature set and the accuracy evaluation result corresponding to the (i-1)-th time-series physiological feature set.

[0050] In operation S140 , when the i-th indicator performance change amount is less than the i−1-th indicator performance change amount, a target physiological feature set is determined from the 1st to i-th time-series physiological feature sets.

[0051] Optionally, the performance change of the i-th indicator is less than the performance change of the i-1-th indicator, which may indicate that the performance improvement of the thermal comfort prediction model tends to decline or saturate, and the gain is in a fluctuating state. The target physiological feature set is the optimal physiological feature set obtained by screening after considering the balance between the simplification of the time-series physiological feature data and the model performance.

[0052] Optionally, after the i-1th time-series physiological feature set is determined as the target physiological feature set, the thermal comfort prediction model is stopped from being used to predict the thermal comfort of the remaining i i time-series physiological feature sets, thereby reducing model overhead.

[0053] Optionally, the target physiological feature set includes time-series physiological feature data corresponding to at least one target feature indicator.

[0054] In operation S150 , a mapping relationship between the group type of the sampling object and the target physiological feature set is established to obtain an object feature mapping relationship table.

[0055] Optionally, the sampled subject group type may be a healthy adult group, a healthy youth group, or other types. Based on the mapping relationship between the sampled subject group type and the target physiological feature set, an object feature mapping relationship table is constructed. The object feature mapping relationship table records the correspondence between the group type and the target physiological feature set under the same target environment.

[0056] Optionally, due to the similarity between the time-series physiological characteristic data of each characteristic indicator and the thermal comfort time-series label, multiple time-series physiological characteristic data are combined to obtain I time-series physiological characteristic sets, and then the difference between the performance change of the i-th indicator and the performance change of the i-1-th indicator is compared. When the i-1-th time-series physiological characteristic set is determined as the target physiological characteristic set, the thermal comfort prediction of the remaining Ii time-series physiological characteristic sets is stopped. In this way, when facing a large number of time-series physiological characteristics with complex structures, efficient feature selection can be achieved while solving the technical problem that the thermal comfort prediction model needs to process all target time-series physiological characteristic data, resulting in large model overhead. The construction of the object feature mapping relationship table helps objects in the same target environment to directly rely on a small amount of time-series physiological characteristic data corresponding to low-redundancy target characteristic indicators for thermal comfort prediction, effectively capture non-steady-state response characteristics, improve the modeling expression ability and prediction accuracy of the thermal comfort response process driven by coupled disturbances, realize accurate prediction of the dynamic evolution process of the thermal comfort state in a continuous time period, and effectively improve the adaptability and computational efficiency of the prediction model.

[0057] Optionally, determining the change in the i-th indicator performance between the i-th time-series physiological feature set and the i-1-th time-series physiological feature set includes: using a prediction model to process the time-series physiological feature data in the i-th time-series physiological feature set to obtain the i-th thermal comfort result; using a performance evaluation function to process the i-th thermal comfort result and the thermal comfort time-series label to obtain the i-th indicator performance; using a prediction model to process the time-series physiological feature data in the i-1-th time-series physiological feature set to obtain the i-1-th thermal comfort result; using a performance evaluation function to process the i-1-th thermal comfort result and the thermal comfort time-series label to obtain the i-1-th indicator performance; and obtaining the i-th indicator performance change based on the i-th indicator performance and the i-1-th indicator performance.

[0058] Optionally, the prediction model is used to process the time series physiological feature data in the first time series physiological feature set to obtain a first thermal comfort result. The performance evaluation function is used to process the first thermal comfort result and the thermal comfort time series label to obtain a first indicator performance.

[0059] Optionally, the prediction model is used to process the time-series physiological feature data in the second time-series physiological feature set to obtain a second thermal comfort result. The second thermal comfort result and the thermal comfort time-series label are processed using a performance evaluation function to obtain a second performance indicator. A change in the second performance indicator is obtained based on the difference between the second performance indicator and the first performance indicator.

[0060] Optionally, the prediction model is used to process the time series physiological feature data in the i-1th time series physiological feature set to obtain the i-1th thermal comfort result. The performance evaluation function is used to process the i-1th thermal comfort result and the thermal comfort time series label to obtain the i-1th indicator performance.

[0061] Optionally, the prediction model is used to process the time-series physiological feature data in the i-th time-series physiological feature set to obtain the i-th thermal comfort result. The i-th thermal comfort result and the thermal comfort time-series label are processed using a performance evaluation function to obtain the i-th indicator performance. The i-th indicator performance change is obtained based on the difference between the i-th indicator performance and the i-1-th indicator performance.

[0062] Optionally, when the i-th thermal comfort result is a continuous variable, the i-th indicator performance can be evaluated using performance evaluation functions such as the Root Mean Squared Error function, the Mean Absolute Error function, the Coefficient of Determination function, and the Mean Absolute Percentage Error function.

[0063] Optionally, when the i-th thermal comfort result is a discrete categorical variable, the i-th indicator performance can be evaluated using performance evaluation functions such as classification accuracy function (Accuracy), precision function (Precision), recall function (Recall), and F1 value function (F1 Score).

[0064] Optionally, a prediction model can be used to process the temporal physiological characteristic data in each temporal physiological characteristic set to achieve continuous prediction and trend identification of thermal comfort across time periods and regions, comprehensively characterizing the dynamic changes in the thermal response of the sampled objects under multi-dimensional perturbations. This overcomes the limitations of traditional static single-point drive control in terms of temporal responsiveness and spatial connectivity, endowing the thermal environment system with the temporal recognition and feedforward control capabilities of dynamic thermal comfort drive, providing key support for the design of zoning control logic for the building thermal environment, full-process comfort tracking, and spatiotemporal coordinated control strategies.

[0065] Optionally, the prediction model includes a gated recurrent unit, a fully connected unit, and a regression unit; wherein, using the prediction model to process the temporal physiological feature data in the i-th temporal physiological feature set to obtain the i-th thermal comfort result includes: using the gated recurrent unit to process the temporal physiological feature data in the i-th temporal physiological feature set to obtain coding features; using the fully connected unit to process the coding features to obtain fully connected features; and using the regression unit to process the fully connected features to obtain the i-th thermal comfort result.

[0066] Optionally, the gated recurrent unit can be constructed based on a gated recurrent layer (GRU).

[0067] Optionally, the temporal physiological feature data in the i-th temporal physiological feature set is input into a gated recurrent unit to output a coding feature. The coding feature is a temporal dependent feature.

[0068] Optionally, the encoded features are input into a fully connected unit to obtain fully connected features, which are features after dimensionality reduction or compression.

[0069] Optionally, the fully connected features are input into a regression unit for classification to obtain the i-th thermal comfort result. The i-th thermal comfort result represents the thermal comfort category to which the prediction belongs. For example, a thermal sensation of 0 indicates a "moderate" category.

[0070] Figure 2 An example diagram of a prediction model according to an embodiment of the present invention is shown.

[0071] like Figure 2 As shown in the figure, the prediction model consists of an input layer, three gated recurrent units, two fully connected units, and a regression unit. The first set of time-series physiological features only includes heart rate data within a certain time window. The number of physiological features in the input layer is 1, and the input data is heart rate data containing 39 sites. The first gated recurrent unit contains 150 hidden layers and outputs an encoded feature at each time step; the second gated recurrent unit contains 200 hidden layers and also outputs an encoded feature at each time step; the third gated recurrent unit contains 75 hidden layers and only returns the output of the last time step. After the three gated recurrent units, there are two fully connected units with 50 and 20 hidden layers, respectively. Finally, the regression unit is used to output the first thermal comfort result, which is a sequence of thermal comfort categories containing 20 sites.

[0072] Optionally, I characteristic indicators are sorted according to the similarity between the time-series physiological characteristic data and the thermal comfort time-series label for the characteristic indicators of the sampling object in the target environment, and the sorting result includes: for each characteristic indicator in the I characteristic indicators, determining the sequence matching result between the time-series physiological characteristic data corresponding to the characteristic indicator and the thermal comfort time-series label, the sequence matching result characterizing the degree of matching between the first sequence length of the time-series physiological characteristic data and the second sequence length of the thermal comfort time-series label; determining the similarity algorithm according to the sequence matching result; calculating the similarity between the time-series physiological characteristic data and the thermal comfort time-series label corresponding to the characteristic indicator using the similarity algorithm to obtain the similarity; and sorting the I characteristic indicators in descending order based on the I similarities to obtain the sorting result.

[0073] Optionally, after preliminarily sorting the time series physiological characteristic data of the characteristic indicators and the thermal comfort time series labels and removing outliers, the sequence matching results of the two in terms of time alignment, sampling consistency, data integrity, structural comparability, etc. are analyzed.

[0074] For example, the time series physiological characteristic data of a characteristic indicator is obtained by sampling 39 times within a time window, so the first sequence length of the time series physiological characteristic data is 39; the thermal comfort time series label is obtained by sampling 20 scores in a questionnaire survey within this time window, so the second sequence length of the thermal comfort time series label is 20, and the sequence matching results of the first sequence length and the second sequence length are inconsistent.

[0075] Optionally, a similarity algorithm corresponding to the sequence matching result is selected, and for the time-series physiological characteristic data of each characteristic indicator, the similarity between the time-series physiological characteristic data of each characteristic indicator and the thermal comfort time-series label is calculated using the similarity algorithm to obtain the similarity corresponding to this characteristic indicator.

[0076] Optionally, multiple similarities are obtained for multiple feature indices, and based on the multiple similarities, the multiple feature indices are sorted in descending order to obtain a sorting result.

[0077] In one embodiment, the similarity As shown in formula (1):

[0078] (1);

[0079] in, The temporal physiological characteristic data representing the jth characteristic index, ,in, represents the observed value of the time series physiological characteristic data of the jth characteristic index at the tth time step, d represents the dimension of the time series physiological characteristic data, Characterize the thermal comfort time series label, Y , Characterizes the score of the thermal comfort time series label at the tth time step, Representation similarity algorithm.

[0080] Optionally, determining the similarity algorithm based on the sequence matching result includes: when the sequence matching result indicates that the length of the first sequence and the length of the second sequence are consistent, determining the similarity algorithm to be a point-by-point matching algorithm; when the sequence matching result indicates that the length of the first sequence and the length of the second sequence are inconsistent, determining the similarity algorithm to be a dynamic time adjustment algorithm.

[0081] Optionally, when the time axes of the time-series physiological characteristic data and the thermal comfort time-series labels are strictly aligned, the sampling frequencies are consistent, and the lengths are equal, and the sequence matching result is determined to be that the length of the first sequence is consistent with the length of the second sequence, algorithms such as Euclidean distance and Manhattan distance are used to process the point-by-point matching relationship between the sequence data.

[0082] Optionally, when there is a time misalignment, sampling rate difference or nonlinear structure mismatch between the time series physiological characteristic data and the thermal comfort time series label, and the sequence matching result is determined to be inconsistent between the length of the first sequence and the length of the second sequence, algorithms such as Dynamic Time Warping (DTW), Fast DTW, and Derivative DTW are used to handle the time axis misalignment or structure mismatch between the sequence data.

[0083] Optionally, when the i-th indicator performance change is less than the i-1th indicator performance change, determining the target physiological feature set from the i-th time-series physiological feature sets includes: when the i-th indicator performance change is less than a first preset threshold, determining the i-1th time-series physiological feature set as the target physiological feature set; when the i-th indicator performance change is less than the i-1th indicator performance change and the i-th indicator performance change is greater than or equal to the first preset threshold, determining the maximum value from the i indicator performance changes; when the maximum value is greater than a second preset threshold, determining the time-series physiological feature set corresponding to the maximum value as the target physiological feature set.

[0084] Optionally, the change in the performance of the i-th indicator is less than a first preset threshold, which may represent that the model performance benefit gain of the prediction model is in a fluctuating state.

[0085] Optionally, the performance change of the i-th indicator is less than the performance change of the i-1-th indicator, which means that with the input of the temporal physiological characteristics of multiple characteristic indicators, the performance benefit of the model begins to show marginal decrease. At this time, the performance change of the i-th indicator is greater than or equal to the first preset threshold, which means that the performance benefit of the model may still be in a large growth state but the growth rate is decreasing.

[0086] Optionally, the second preset threshold is greater than the first preset threshold, and a maximum value is determined from the i indicator performance changes, with the maximum value being the highest point of growth. If the maximum value is greater than the second preset threshold, the time-series physiological feature set corresponding to the maximum value is determined as the target physiological feature set.

[0087] Optionally, a performance-complexity curve can be constructed based on the performance of the feature indicators of each temporal physiological feature set and the number of feature indicators included in the temporal physiological feature set. This curve quantifies the global response of feature indicator performance to the number of feature indicators. Feature indicators are gradually introduced until an optimal trade-off point is identified. The optimal trade-off point is the turning point where the marginal improvement in feature indicator performance decreases with increasing the number of feature indicators, while the number of feature indicators increases significantly. This point represents the optimal balance between prediction accuracy and computational complexity, enabling the model to be lightweight and efficient while maintaining accuracy. To identify this inflection point, model selection criteria such as the elbow method, performance-to-complexity ratio, and marginal performance gain analysis can be introduced to determine the target temporal physiological feature set with optimal information contribution and minimal redundancy.

[0088] Optionally, the high-dimensional redundancy and structural complexity of data input in existing methods are key challenges that limit the lightweight construction of thermal comfort models and the adaptation of intelligent monitoring equipment. The present invention is based on the feature lightweight modeling process of time series similarity sorting and combination, combined with the performance-complexity joint evaluation, which can screen out a set of target time series physiological features with high information contribution and low redundancy, while maintaining the prediction accuracy, effectively reducing the model input dimension and structural complexity. It makes up for the shortcomings of traditional modeling methods in feature redundancy control and system adaptability, provides an efficient modeling path for multi-source heterofrequency feature acquisition, embedded deployment and thermal comfort prediction under edge operating conditions, and helps the lightweight integration and practical application of multi-source heterofrequency time series intelligent monitoring equipment.

[0089] Figure 3 An example diagram showing the division of different functional spaces in a full-process simulation experiment of a high-speed railway station ride according to an embodiment of the present invention is shown.

[0090] like Figure 3As shown, the target environment can be a large high-speed rail station in a cold region. Sixteen healthy adult passengers were sampled and a full-process high-speed rail station ride simulation experiment was conducted in an artificial climate chamber within the variable space and environmental experimental platform (a). This experiment simulated the complex, time-varying characteristics of thermal comfort, triggered by coupled changes in environmental factors and metabolic levels throughout the passenger journey. Information on environmental factors and passenger metabolic levels was obtained for the various functional spaces within this large high-speed rail station during the winter. Based on the survey results, the full journey for passengers entering the station by public transportation includes five functional spaces: outdoor entry space O, security check space A, travel space B, waiting space C, and ticket checking space D. The adaptation space P, used by the 16 healthy adult passengers for pre-experiment rest, briefing on the experimental procedures, and changing clothing, was primarily used for quiet sitting before the experiment to reduce the effects of short-term thermal exposure. All sampled subjects wore uniform clothing throughout the experiment. The behavior pattern of healthy adult passengers in the adaptation space P is sitting still for 30 minutes, the behavior pattern of healthy adult passengers in the outdoor entry space O is walking for 2 minutes and 40 seconds at a walking speed of 1.32 m / s, the behavior pattern of healthy adult passengers in the security check space A is walking for 1 minute and 55 seconds at a walking speed of 0.55 m / s, the behavior pattern of healthy adult passengers in the travel space B is walking for 1 minute at a walking speed of 1.32 m / s, the behavior pattern of healthy adult passengers in the waiting space C is sitting still for 23 minutes and 30 seconds, and the behavior pattern of healthy adult passengers in the ticket checking space D is standing for 8 minutes and 30 seconds.

[0091] Optionally, in a simulation experiment, the temperature of each functional space and the metabolic level of the sampled object are controlled to induce changes in the temporal physiological characteristic data and thermal comfort temporal labels of the sampled object. The temperature of different functional spaces is controlled by the radiant floor covering the ground and the lateral fan coil units, while the other environmental parameters are kept constant. The change in individual metabolic level is mainly controlled by the change in behavioral patterns, which include sitting, standing and walking. The walking speed is controlled by evenly spaced points on the ground and a metronome.

[0092] Figure 4 An example diagram of obtaining simulation experiment data according to an embodiment of the present invention is shown.

[0093] like Figure 4As shown in the figure, in the simulation experiment, time-series physiological characteristic data and thermal comfort time-series labels were collected for each characteristic indicator of the sampled subjects at each experimental stage under different temperature sequence conditions within a complete time window. The characteristic indicators include heart rate characteristics and skin temperature characteristics. Skin temperature characteristics include eight characteristic indicators: forehead skin temperature, chest skin temperature, forearm skin temperature, hand back skin temperature, thigh skin temperature, calf skin temperature, foot back skin temperature, and average skin temperature. The average skin temperature characteristic was calculated using a seven-point calculation method. Thermal comfort time-series labels used Thermal Sensation Vote (TSV) labels and Thermal Satisfaction Vote labels. The labels were unevenly sampled at each stage through questionnaires. The TSV labels were a continuous scale of -4 to 4, where the values ​​"-4," "-3," "-2," "-1," "0," "1," "2," "3," and "4" represent "very cold," "cold," "cool," "slightly cool," "moderate," "slightly warm," "warm," "hot," and "very hot," respectively. The thermal satisfaction vote labels were "satisfied," "fair," and "unsatisfied." The simulation experiment includes 7 experimental stages. In the first experimental stage, the sampling subjects prepared for the experiment in the adaptation space P for 10 minutes. In the second experimental stage, the sampling subjects sat quietly in the adaptation space P for 30 minutes and started to collect time-series physiological characteristic data. In the third experimental stage, the sampling subjects walked in the outdoor entry space O for 2 minutes and 40 seconds at a walking speed of 1.32 m / s and started to make thermal sensation voting labels and thermal satisfaction voting labels. In the fourth experimental stage, the sampling subjects walked in the security check space A for 1 minute and 15 seconds and then stood for 40 seconds at a walking speed of 0.55 m / s. In the fifth experimental stage, the sampling subjects walked in the moving space B for 2 minutes at a walking speed of 1.32 m / s. In the sixth experimental stage, the sampling subjects sat quietly in the waiting space C for 23 minutes and 30 seconds. In the seventh experimental stage, the sampling subjects stood in the ticket checking space D for 8 minutes and 30 seconds.

[0094] Alternatively, the time-series physiological feature data for skin temperature and heart rate features is collected at a one-minute interval, while the thermal comfort time-series labels are unevenly sampled. A complete working condition generates 39 sets of time-series physiological feature data and 20 sets of thermal comfort time-series labels, resulting in misaligned data granularity. Therefore, a dynamic time adjustment algorithm is used as a similarity function to align the time-series physiological feature data and thermal comfort time-series labels and calculate similarities. Multiple similarities are obtained, and the nine feature indicators are ranked by time series similarity.

[0095] Table 1 shows a similarity table between the time-series physiological characteristic data of each characteristic index and the thermal comfort time-series label according to an embodiment of the present invention.

[0096] Table 1 Similarity

[0097]

[0098] Optionally, as shown in Table 1, the sorting results are obtained by sorting from high to low according to the similarity of the feature indicators, which are heart rate features, back of hand skin temperature features, thigh skin temperature features, calf skin temperature features, instep skin temperature features, average skin temperature features, forearm skin temperature features, forehead skin temperature features, and chest skin temperature features.

[0099] Table 2 shows a time series physiological feature set table according to an embodiment of the present invention.

[0100] Table 2 Time series physiological feature set

[0101]

[0102] Optionally, the time-series physiological characteristic data for each of the multiple characteristic indicators is heart rate data, hand back skin temperature data, thigh skin temperature data, calf skin temperature data, instep skin temperature data, average skin temperature data, forearm skin temperature data, forehead skin temperature data, and chest skin temperature data. The multiple time-series physiological characteristic data are combined based on the sorting results and preset combination rules. As shown in Table 2, based on the sorting results, the time-series physiological characteristic data are added one by one, generating a total of nine time-series physiological characteristic sets (F1, F2, F3, F4, F5, F6, F7, F8, and F9).

[0103] Optionally, a thermal comfort result is generated for each time series physiological feature set, and based on the thermal comfort result and the thermal comfort time series label, the classification prediction accuracy of the model under the time series physiological feature set is calculated as an indicator performance.

[0104] In one embodiment, the classification prediction accuracy is shown in formula (2):

[0105] (2);

[0106] Where T represents the number of predictions; represents the score of the thermal comfort time series label at the tth time step; represents the predicted value of thermal comfort result at time step t; represents the indicator function, .

[0107] Optionally, when the index performance of the i-th temporal physiological feature set is obtained, the i-th index performance change between the i-th temporal physiological feature set and the i-1-th temporal physiological feature set is calculated, and the marginal performance gain is calculated based on the i-th index performance change.

[0108] Table 3 shows an indicator performance change table according to an embodiment of the present invention.

[0109] Table 3 Performance changes of indicators

[0110]

[0111] Alternatively, time-series physiological feature set F3 in Table 3 is where the accuracy first shows a significant jump, with a 13% change in feature index performance. Subsequently, the model's index performance changes fluctuated within a ±3% range, indicating saturation of performance improvement. Therefore, time-series physiological feature set F3 was determined as the target time-series physiological feature set, and the target time-series physiological feature data was determined to be: heart rate data, back of hand skin temperature data, and thigh skin temperature data.

[0112] Figure 5 A flow chart of a thermal comfort prediction method based on temporal physiological characteristics according to an embodiment of the present invention is shown.

[0113] like Figure 5 As shown, the thermal comfort prediction method 500 based on time-series physiological characteristics includes operations S510 to S530.

[0114] In operation S510 , in response to a thermal comfort prediction request for a target object located in a target environment, a target object group type to which the target object belongs is determined.

[0115] In operation S520 , a target feature index corresponding to the target object is determined according to the object feature mapping relationship table and the target object group type.

[0116] In operation S530 , the prediction model is used to process the time-series physiological characteristic data of the target characteristic index corresponding to the target object to obtain a thermal comfort prediction result of the target object.

[0117] Optionally, the object feature mapping relationship table is a correspondence between sampling objects belonging to different object group types and target feature indicators. The object feature mapping relationship table is obtained according to the above-mentioned method for determining temporal physiological characteristics under the target environment.

[0118] Optionally, in response to a thermal comfort prediction request for a target object in a target environment, a target object group type to which the target object belongs is determined. For example, the target object group type of the target object is an adult healthy object group type.

[0119] Optionally, the target physiological feature set corresponding to the healthy object group type is determined from the object feature mapping relationship table to be the time series physiological feature set F3. Therefore, the target feature indicators corresponding to the target object are determined to be the heart rate feature, the back of the hand skin temperature feature, and the thigh skin temperature feature.

[0120] Alternatively, the target subject's heart rate data, hand and thigh skin temperature data can be directly obtained and processed using a prediction model to obtain a thermal comfort prediction result for the target subject. The thermal comfort prediction result can be a thermal sensation of 0, where 0 represents a "moderate" category.

[0121] Optionally, the prediction model can be a dynamic thermal comfort prediction model based on a sequence-to-sequence architecture. The prediction model can use a recurrent neural network (RNN) as a basic unit and can integrate optimized structures such as a long short-term memory network (Long Short-Term Memory Network), a gated recurrent unit (GRU), and a bidirectional recurrent neural network (Bidirectional Recurrent Neural Network).

[0122] Optionally, in the target environment, according to the target object group type to which the target object belongs, a small number of target feature indicators with strong correlation that have a mapping relationship with the target object group type are selected from the object feature mapping relationship table. When thermal comfort prediction is performed directly based on the time-series physiological feature data of the target feature indicators of the target object, there is no need to obtain the time-series physiological feature data of all feature indicators, which improves the prediction accuracy while reducing the model overhead.

[0123] Based on the above method for determining temporal physiological characteristics under target environment, the present invention also provides a device for determining temporal physiological characteristics under target environment. Figure 6 The device is described in detail.

[0124] Figure 6 A structural block diagram of a device for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown.

[0125] like Figure 6 As shown, the device 600 for determining temporal physiological characteristics in a target environment of this embodiment includes a sorting module 610 , a combining module 620 , a first determining module 630 , a second determining module 640 and a mapping module 650 .

[0126] Sorting module 610 is configured to sort each of the I characteristic indicators based on the similarity between the time-series physiological characteristic data of the sampled subject in the target environment and the thermal comfort time-series label for the characteristic indicator, thereby obtaining a sorting result. In one embodiment, sorting module 610 may be configured to perform operation S110 described above, and will not be further described herein.

[0127] The combining module 620 is configured to combine the plurality of time-series physiological characteristic data based on a preset combining rule and a sorting result to obtain I time-series physiological characteristic sets, where I is an integer greater than 1. In one embodiment, the combining module 620 may be configured to perform the operation S120 described above, which will not be described in detail here.

[0128] The first determining module 630 is configured to determine the i-th indicator performance change between the i-th time-series physiological feature set and the i-1-th time-series physiological feature set. The i-th indicator performance change represents the degree of change between the accuracy of assessing the thermal comfort of the sampled subject based on the i-1th time-series physiological feature set and the accuracy of assessing the thermal comfort of the sampled subject based on the i-th time-series physiological feature set. The i-th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-th position in the sorted result, and the i-1th time-series physiological feature set includes the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-1th position in the sorted result. In one embodiment, the first determination module 630 can be used to perform operation S130 described above, which will not be repeated here.

[0129] The second determination module 640 is configured to determine a target physiological feature set from the first through the i-th time-series physiological feature sets when the change in the i-th indicator performance is less than the change in the i-1-th indicator performance, where the target physiological feature set includes time-series physiological feature data corresponding to at least one target feature indicator. In one embodiment, the second determination module 640 may be configured to perform operation S140 described above, which is not further described here.

[0130] The mapping module 650 is used to establish a mapping relationship between the group type of the sampled objects and the target physiological feature set to obtain an object feature mapping relationship table. In one embodiment, the mapping module 650 can be used to perform the operation S150 described above, which will not be repeated here.

[0131] Optionally, the first determining module 630 includes a first determining submodule, a second determining submodule, a third determining submodule, a fourth determining submodule, and a fifth determining submodule.

[0132] The first determination submodule is used to process the time series physiological feature data in the i-th time series physiological feature set using the prediction model to obtain the i-th thermal comfort result.

[0133] The second determination submodule is used to process the i-th thermal comfort result and the thermal comfort time series label using the performance evaluation function to obtain the i-th indicator performance.

[0134] The third determination submodule is used to process the time series physiological feature data in the i-1th time series physiological feature set by using the prediction model to obtain the i-1th thermal comfort result.

[0135] The fourth determination submodule is used to process the i-1th thermal comfort result and the thermal comfort time series label using the performance evaluation function to obtain the i-1th indicator performance.

[0136] The fifth determining submodule is used to obtain the change in the i-th indicator performance according to the i-th indicator performance and the (i-1)-th indicator performance.

[0137] Figure 7 The figure shows a structural block diagram of a thermal comfort prediction device based on time series physiological characteristics according to an embodiment of the present invention.

[0138] like Figure 7 As shown, the thermal comfort prediction device 700 based on time-series physiological characteristics of this embodiment includes a response module 710 , a third determination module 720 and a prediction module 730 .

[0139] The response module 710 is configured to respond to the thermal comfort prediction request for the target object in the target environment and determine the target object group type to which the target object belongs. In one embodiment, the response module 710 may be configured to execute the operation S510 described above, which will not be described in detail here.

[0140] A third determination module 720 is configured to determine a target feature indicator corresponding to the target object based on the object feature mapping relationship table and the target object group type. The object feature mapping relationship table represents the correspondence between sampled objects belonging to different object group types and the target feature indicator. The object feature mapping relationship table is obtained according to the method for determining temporal physiological features under the target environment. In one embodiment, the third determination module 720 can be configured to perform operation S520 described above, and will not be further described here.

[0141] The prediction module 730 is configured to process the time series physiological characteristic data of the target characteristic index corresponding to the target object using the prediction model to obtain a thermal comfort prediction result for the target object. In one embodiment, the prediction module 730 can be configured to perform the operation S530 described above, which will not be described in detail here.

[0142] Optionally, any multiple modules among the modules, submodules, units, and subunits may be combined into a single module, or any one of the modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. At least one of the modules, submodules, units, and subunits may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware through any other reasonable means of circuit integration or packaging, or implemented in any one of the three implementation methods of software, hardware, and firmware, or any appropriate combination of any of these. Alternatively, at least one of the modules, submodules, units, and subunits may be at least partially implemented as a computer program module that, when executed, performs the corresponding functionality.

[0143] Figure 8 A block diagram of an electronic device suitable for implementing a method for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention is shown.

[0144] Figure 8 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0145] like Figure 8 As shown, a computer electronic device 800 according to an embodiment of the present invention includes a processor 801, which can perform various appropriate actions and processes according to a program stored in a ROM 802 (read-only memory) or a program loaded from a storage unit 508 into a RAM 803 (random access memory). The processor 801 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 801 may also include onboard memory for caching purposes. The processor 801 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0146] The RAM 803 stores various programs and data required for the operation of the electronic device 800. The processor 801, ROM 802, and RAM 803 are connected to each other via a bus 804. The processor 801 executes the programs in the ROM 802 and / or RAM 803 to perform the various operations of the method flow according to the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than the ROM 802 and RAM 803. The processor 801 may also execute the programs stored in one or more memories to perform the various operations of the method flow according to the embodiment of the present invention.

[0147] Optionally, electronic device 800 may further include an input / output (I / O) interface 805, which is also connected to bus 804. Electronic device 800 may also include one or more of the following components connected to I / O interface 805: an input section 806 including a keyboard, mouse, etc.; an output section 807 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 808 including a hard disk; and a communication section 809 including a network interface card such as a LAN card or modem. Communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to I / O interface 805 as needed. Removable media 811, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 810 as needed, so that computer programs read from the removable media can be installed into storage section 808 as needed.

[0148] Optionally, the method flow according to an embodiment of the present invention can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 809, and / or installed from the removable medium 811. When the computer program is executed by the processor 801, the above-mentioned functions defined in the system of the embodiment of the present invention are performed. Optionally, the system, device, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0149] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method for determining temporal physiological characteristics in a target environment according to an embodiment of the present invention.

[0150] Optionally, the computer-readable storage medium may be a non-volatile computer-readable storage medium. Examples include, but are not limited to, a portable computer disk, a hard disk, random access memory (RAM 803), read-only memory (ROM 802), erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0151] For example, optionally, the computer-readable storage medium may include the ROM 802 and / or RAM 803 described above and / or one or more memories other than the ROM 802 and RAM 803 .

[0152] An embodiment of the present invention also includes a computer program product, which includes a computer program, which contains program code for executing the method provided by the embodiment of the present invention. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the method for determining temporal physiological characteristics in a target environment provided by the embodiment of the present invention.

[0153] When the computer program is executed by the processor 801, the above functions defined in the system / device of the embodiment of the present invention are performed. Optionally, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0154] In one embodiment, the computer program may be stored on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal on a network medium, downloaded and installed via the communication portion 809, and / or installed from a removable medium 811. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0155] Optionally, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, Python, "C" language, or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0156] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes may occur in an order different from that marked in the accompanying drawings. For example, two boxes shown in succession may actually be executed substantially in parallel, or they may sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, as well as the combination of boxes in the block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or may be implemented using a combination of dedicated hardware and computer instructions. It will be understood by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or combinations fall within the scope of the present invention.

[0157] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A method for determining temporal physiological characteristics in a target environment, characterized in that: The method comprises: For each of the I characteristic indicators, the I characteristic indicators are sorted according to the similarity between the time-series physiological characteristic data of the sampling subject for the characteristic indicator in the target environment and the thermal comfort time-series label to obtain a sorting result; Based on a preset combination rule and the sorting result, the plurality of time-series physiological characteristic data are combined to obtain I time-series physiological characteristic set, where I is an integer greater than 1; Determine the change in performance of the i-th indicator between the i-th temporal physiological feature set and the i-1-th temporal physiological feature set, The i-th indicator performance change represents a degree of change between the accuracy of evaluating the thermal comfort of the sampled subject based on the i-1th time-series physiological feature set and the accuracy of evaluating the thermal comfort of the sampled subject based on the i-th time-series physiological feature set, the i-th time-series physiological feature set including the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-th position in the sorting result, and the i-1th time-series physiological feature set including the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-1th position in the sorting result; When the change in the performance of the i-th indicator is less than the change in the performance of the i-1-th indicator, determining a target physiological feature set from the first to the i-th time-series physiological feature sets, wherein the target physiological feature set includes time-series physiological feature data corresponding to at least one target feature indicator; A mapping relationship between the group type of the sampling object and the target physiological feature set is established to obtain an object feature mapping relationship table.

2. The method according to claim 1, characterized in that The determining of the i-th indicator performance change between the i-th time-series physiological feature set and the i-1-th time-series physiological feature set includes: Processing the time series physiological characteristic data in the i-th time series physiological characteristic set using the prediction model to obtain an i-th thermal comfort result; Processing the i-th thermal comfort result and the thermal comfort time series label using a performance evaluation function to obtain an i-th indicator performance; Processing the time series physiological feature data in the (i-1)th time series physiological feature set using the prediction model to obtain an (i-1)th thermal comfort result; Processing the (i-1)th thermal comfort result and the thermal comfort time series label using a performance evaluation function to obtain an (i-1)th indicator performance; The change in the i-th indicator performance is obtained according to the i-th indicator performance and the i-1-th indicator performance.

3. The method according to claim 2, characterized in that The prediction model includes a gated recurrent unit, a fully connected unit, and a regression unit; The step of processing the time series physiological feature data in the i-th time series physiological feature set using a prediction model to obtain the i-th thermal comfort result includes: Processing the temporal physiological feature data in the i-th temporal physiological feature set using the gated recurrent unit to obtain a coding feature; Processing the encoding feature using the fully connected unit to obtain a fully connected feature; The fully connected features are processed using the regression unit to obtain the i-th thermal comfort result.

4. The method according to claim 1, wherein The characteristic indicators are sorted according to the similarity between the time series physiological characteristic data of the characteristic indicators and the thermal comfort time series label of the sampling object in the target environment, and the sorting results include: For each characteristic indicator of the I characteristic indicators, determining a sequence matching result between the time-series physiological characteristic data corresponding to the characteristic indicator and the thermal comfort time-series label, wherein the sequence matching result represents a degree of matching between a first sequence length of the time-series physiological characteristic data and a second sequence length of the thermal comfort time-series label; Determining a similarity algorithm based on the sequence matching result; Calculating the similarity between the time series physiological characteristic data corresponding to the characteristic index and the thermal comfort time series label using the similarity algorithm to obtain a similarity; Based on the I similarities, the I feature indicators are sorted in descending order to obtain the sorting result.

5. The method according to claim 4, characterized in that Determining a similarity algorithm according to the sequence matching result includes: When the sequence matching result indicates that the length of the first sequence is consistent with the length of the second sequence, determining that the similarity algorithm is a point-by-point matching algorithm; When the sequence matching result indicates that the length of the first sequence is inconsistent with the length of the second sequence, the similarity algorithm is determined to be a dynamic time adjustment algorithm.

6. The method according to claim 1, characterized in that When the i-th indicator performance change is less than the i-1-th indicator performance change, determining the target physiological feature set from the i-th time-series physiological feature sets includes: When the change in the i-th indicator performance is less than a first preset threshold, determining the i-1th time-series physiological feature set as the target physiological feature set; When the i-th indicator performance change amount is less than the i-1-th indicator performance change amount, and the i-th indicator performance change amount is greater than or equal to the first preset threshold, determining a maximum value from the i indicator performance change amounts; In a case where the maximum value is greater than a second preset threshold, the time-series physiological feature set corresponding to the maximum value is determined as the target physiological feature set.

7. A thermal comfort prediction method based on temporal physiological characteristics, characterized in that: The method comprises: In response to a thermal comfort prediction request for a target object in a target environment, determining a target object group type to which the target object belongs; Determining a target feature index corresponding to the target object according to an object feature mapping relationship table and the target object group type, wherein the object feature mapping relationship table represents a correspondence between sampled objects belonging to different object group types and target feature indexes, and the object feature mapping relationship table is obtained by the method according to any one of claims 1 to 6; The prediction model is used to process the time series physiological characteristic data of the target characteristic index corresponding to the target object to obtain a thermal comfort prediction result of the target object.

8. A device for determining temporal physiological characteristics in a target environment, characterized in that: The device comprises: a sorting module, configured to sort each of the I characteristic indicators according to a similarity between the time-series physiological characteristic data of the sampling subject in the target environment and the thermal comfort time-series label for the characteristic indicator, thereby obtaining a sorting result; a combining module, configured to combine the plurality of time-series physiological characteristic data based on a preset combination rule and the sorting result to obtain I time-series physiological characteristic set, where I is an integer greater than 1; The first determining module is used to determine the i-th indicator performance change between the i-th time-series physiological feature set and the i-1-th time-series physiological feature set. , the i-th indicator performance change represents a degree of change in the accuracy of evaluating the thermal comfort of the sampled subject based on the i-th time-series physiological feature set compared to the i-1-th time-series physiological feature set, the i-th time-series physiological feature set including the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-th position in the sorting result, and the i-1-th time-series physiological feature set including the time-series physiological feature data corresponding to the feature indicators located from the 1st position to the i-1-th position in the sorting result; a second determining module, configured to determine a target physiological feature set from the i time-series physiological feature sets when the i-th indicator performance change is less than the i-1-th indicator performance change, wherein the target physiological feature set includes time-series physiological feature data corresponding to at least one target feature indicator; The mapping module is used to establish a mapping relationship between the group type of the sampling object and the target physiological feature set to obtain an object feature mapping relationship table.

9. A thermal comfort prediction device based on temporal physiological characteristics, characterized in that: The device comprises: A response module, configured to determine, in response to a thermal comfort prediction request for a target object located in a target environment, a target object group type to which the target object belongs; a third determining module, configured to determine a target feature indicator corresponding to the target object based on an object feature mapping relationship table and the target object group type, wherein the object feature mapping relationship table is a correspondence between sampled objects belonging to different object group types and target feature indicators, and the object feature mapping relationship table is obtained by the method according to any one of claims 1 to 6; The prediction module is used to process the time-series physiological characteristic data of the target characteristic index corresponding to the target object using the prediction model to obtain the thermal comfort prediction result of the target object.

10. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7.

Images