Building thermal parameter decoupling identification method, device, equipment and medium

Abstract

This invention relates to the field of materials testing technology, addressing the problem of low accuracy in identifying thermal parameters of building envelopes in existing technologies. It provides a decoupled identification method, apparatus, equipment, and medium for thermal parameters of building envelopes. The method includes: performing an equivalent model of the one-dimensional unsteady-state heat transfer process of the building envelope under test to obtain a parameterized dynamic heat transfer model; constructing a Fischer information matrix based on the parameterized dynamic heat transfer model; using the corresponding objective function value as the optimization objective to solve for the input signal of the wind tunnel experimental environment to obtain the target dynamic excitation signal; conducting experimental tests on the building envelope under the action of the target dynamic excitation signal, and updating the Fischer information matrix based on the experimental data; when the experimental termination condition is met according to the updated Fischer information matrix, obtaining the target thermal parameters based on the acquired target data and the parameterized dynamic heat transfer model. This invention improves the identification accuracy of key thermal parameters of building envelopes.

Description

Technical Field

[0001] This invention relates to the field of materials testing technology, and in particular to a method, apparatus, equipment and medium for decoupling and identifying thermal parameters of building envelope. Background Technology

[0002] Chinese invention patent application CN101556255A, published on October 14, 2009, discloses a method for analyzing on-site thermal resistance test data of building envelopes. While considering the influence of wind speed on the thermal resistance test results, it does not consider the influence of material arrangement on thermal performance. Furthermore, existing steady-state testing methods require separate testing of thermal resistance and heat capacity, resulting in long testing cycles and the inability to distinguish the effects of thermal resistance and heat capacity during the testing process. Especially under natural weather conditions or simple sinusoidal wave excitation, the effects of thermal resistance and heat capacity on the heat flow signal highly overlap, leading to uncertainty and multiple solutions in the experimental results.

[0003] Therefore, there is an urgent need for an experimental method that can accurately and efficiently decouple and identify thermal parameters in a short time, in order to solve the problem of low accuracy in identifying thermal parameters of building envelopes in existing technologies. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, device and medium for decoupling and identifying thermal parameters of building envelope, in order to solve the problem of low accuracy in identifying thermal parameters of building envelope in the prior art.

[0005] In a first aspect, embodiments of the present invention provide a method for decoupling and identifying thermal parameters of an enclosure structure, the method comprising:

[0006] An equivalent model of the one-dimensional unsteady heat transfer process of the building envelope under test is performed to obtain a parameterized dynamic heat transfer model.

[0007] Based on the parameterized dynamic heat transfer model, a Fischer information matrix is ​​constructed;

[0008] Using the objective function value corresponding to the Fischer information matrix as the optimization objective, the wind tunnel experimental environment input signal is solved to obtain the target dynamic excitation signal;

[0009] Experimental tests were conducted on the enclosure structure under the action of the target dynamic excitation signal to obtain experimental data, and the Fischer information matrix was updated based on the obtained experimental data.

[0010] When the preset experimental termination condition is met according to the updated Fischer information matrix, the acquired experimental data is used as the target data.

[0011] Based on the target data and the parameterized dynamic heat transfer model, the thermal parameters to be identified are jointly decoupled and identified to obtain the target thermal parameters.

[0012] In an optional embodiment, the one-dimensional unsteady-state heat transfer process of the enclosure structure under test is equivalently modeled to obtain a parameterized dynamic heat transfer model, including:

[0013] The one-dimensional unsteady heat transfer process is transformed into an equivalent model to obtain a thermal network model;

[0014] The thermal network model is transformed into a state-space form to obtain a standard state-space model.

[0015] The thermal parameters to be identified are associated with the model parameters in the state-space form model to obtain a parameterized dynamic heat transfer model.

[0016] In an optional embodiment, constructing the Fischer information matrix based on the parameterized dynamic heat transfer model includes:

[0017] Based on the parameterized dynamic heat transfer model, the partial derivatives of the model output response with respect to the thermal parameters to be identified are calculated to obtain the parameter sensitivity function value.

[0018] Based on the parameter sensitivity function values ​​at each discrete time point and the covariance matrix of the measurement noise, a Fischer information matrix is ​​constructed.

[0019] In an optional embodiment, the step of using the objective function value corresponding to the Fischer information matrix as the optimization objective to solve for the wind tunnel experimental environment input signal to obtain the target dynamic excitation signal includes:

[0020] The wind tunnel experimental environment input signal over continuous time is subjected to time-domain parameterization to obtain finite-dimensional optimization variables;

[0021] Based on the finite-dimensional optimization variables, a corresponding time series function is constructed, and candidate dynamic excitation signals are generated;

[0022] Based on the preset parameter nominal values, the candidate dynamic excitation signal is input into the parameterized dynamic heat transfer model to determine the candidate output response signal;

[0023] Based on each candidate output response signal, the objective function value of each Fischer information matrix is ​​calculated;

[0024] Compare the objective function values ​​and select the candidate dynamic excitation signal corresponding to the Fischer information matrix with the largest objective function value as the target dynamic excitation signal.

[0025] In an optional embodiment, the step of conducting experimental tests on the enclosure structure under the action of the target dynamic excitation signal, acquiring experimental data, and updating the Fischer information matrix based on the acquired experimental data includes:

[0026] Under the action of the target dynamic excitation signal, the enclosure structure is tested experimentally, and corresponding experimental data are collected at multiple discrete time points;

[0027] Based on the experimental data at each discrete time point and the parameterized dynamic heat transfer model, the sensitivity function value of the target parameter of the thermal parameter to be identified in the model output response is determined.

[0028] The Fischer information matrix is ​​updated based on the target parameter sensitivity function values ​​and the covariance matrix of the measurement noise at each discrete time point.

[0029] In an optional embodiment, the step of using the acquired experimental data as target data when determining that a preset experiment termination condition is met based on the updated Fischer information matrix includes:

[0030] Based on the updated Fischer information matrix, obtain the determinant value of the Fischer information matrix at each moment within the preset time interval;

[0031] The growth rate of the determinant is calculated based on the determinant values ​​of each Fischer information matrix.

[0032] When the growth rate of the determinant is less than the preset growth rate threshold, it is determined that the preset experiment termination condition is met, and the acquired experimental data is used as the target data.

[0033] In an optional embodiment, the step of jointly decoupling and identifying the thermal parameters to be identified based on the target data and the parameterized dynamic heat transfer model to obtain the target thermal parameters includes:

[0034] The target data is associated with the parameterized dynamic heat transfer model to determine the model output response of the parameterized dynamic heat transfer model when the thermal parameter to be identified takes different values.

[0035] Based on the deviation between the model output response and the target data, a parameter optimization problem is constructed.

[0036] The target thermal parameters are obtained by jointly solving the parameter optimization problem using either the maximum likelihood estimation method or the nonlinear least squares method.

[0037] Secondly, embodiments of the present invention provide a device for decoupling and identifying thermal parameters of an enclosure structure, the device comprising:

[0038] The model building module is used to perform equivalent modeling of the one-dimensional unsteady heat transfer process of the building envelope under test, and obtain a parameterized dynamic heat transfer model.

[0039] The matrix construction module is used to construct a Fischer information matrix based on the parameterized dynamic heat transfer model.

[0040] The excitation signal determination module is used to take the objective function value corresponding to the Fischer information matrix as the optimization objective, solve the wind tunnel experimental environment input signal, and obtain the target dynamic excitation signal.

[0041] The matrix update module is used to conduct experimental tests on the enclosure structure under the action of the target dynamic excitation signal, acquire experimental data, and update the Fischer information matrix based on the acquired experimental data.

[0042] The target data determination module is used to take the acquired experimental data as target data when it is determined that the preset experimental termination condition is met according to the updated Fischer information matrix.

[0043] The decoupling module is used to perform joint decoupling identification of the thermal parameters to be identified based on the target data and the parameterized dynamic heat transfer model, so as to obtain the target thermal parameters.

[0044] Thirdly, embodiments of the present invention provide an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method of the first aspect described above.

[0045] Fourthly, embodiments of the present invention provide a storage medium storing computer program instructions, which, when executed by a processor, implement the method of the first aspect described above.

[0046] In summary, the beneficial effects of the present invention are as follows:

[0047] The method, apparatus, equipment, and medium for decoupling and identifying thermal parameters of building envelopes provided in this invention construct a parameterized dynamic heat transfer model, unifying the key thermal parameters of the building envelope within the same framework. A Fischer information matrix is ​​introduced to quantitatively evaluate the identifiability of parameters under different experimental input conditions. The objective function value is used as the optimization target to obtain the dynamic excitation signal with optimal information utilization efficiency, enabling the experimental environment input to specifically stimulate differentiated responses of each parameter, thereby effectively reducing the coupling degree of parameters such as thermal resistance and heat capacity in the time domain. During the experiment, the Fischer information matrix is ​​updated in real time, and the experimental termination condition is determined based on the updated Fischer information matrix to avoid interference from invalid data, ensuring the authenticity and reliability of the acquired target data. Finally, based on this target data and the parameterized dynamic heat transfer model, the thermal parameters to be identified are jointly decoupled and identified, ensuring that the contribution of each parameter to the output response is fully distinguished, reducing ambiguity and correlation, thereby significantly improving the accuracy of identifying thermal parameters of the building envelope. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of the present invention.

[0049] Figure 1 This is a schematic diagram of the overall process of the decoupling and identification method for thermal parameters of the building envelope in Embodiment 1 of the present invention;

[0050] Figure 2 This is a schematic diagram of the optimal target dynamic excitation signal in Embodiment 1 of the present invention;

[0051] Figure 3 This is a schematic diagram of the structure of the thermal parameter decoupling identification device for the building envelope in Embodiment 2 of the present invention;

[0052] Figure 4 This is a schematic diagram of the electronic device in Embodiment 3 of the present invention. Detailed Implementation

[0053] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only configured to explain the present invention and are not configured to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the invention.

[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0055] Example 1

[0056] South China University of Technology, relying on the State Key Laboratory of Subtropical Building Science, has built a wind tunnel experimental platform for research on building environment and wind engineering. This type of experimental platform can dynamically control various environmental parameters such as wind speed, air temperature, solar radiation, and relative humidity, and conduct related thermal performance tests on building envelopes under controlled conditions. The equipment can continuously adjust parameters such as air temperature, humidity, wind speed, and solar radiation to simulate complex climatic conditions such as high temperature, high humidity, and strong radiation, providing an experimental basis for the study of the thermal response of building envelopes under actual service environments.

[0057] In practical wind tunnel experiments, although it is possible to achieve joint control of multiple environmental parameters, conventional experiments often rely on empirically set or simple variation patterns for environmental input. The influence of various thermal parameters in the dynamic response is easily superimposed, making it difficult to effectively distinguish the contributions of different parameters in the test results. Therefore, this embodiment introduces an experimental design method based on the Fischer information matrix, optimizing the wind tunnel environmental input to make the temporal changes of each environmental parameter more targeted, thereby improving the accuracy and efficiency of identifying the thermal resistance, heat capacity, and surface heat transfer parameters of the building envelope.

[0058] Please see Figure 1 This invention provides a method for decoupling and identifying thermal parameters of building envelope, the method comprising:

[0059] S1: Equivalent modeling of the one-dimensional unsteady heat transfer process of the building envelope to be tested is performed to obtain a parameterized dynamic heat transfer model;

[0060] Specifically, the one-dimensional unsteady-state heat transfer process of the building envelope refers to a heat transfer process in which heat is transferred in only one direction, and the temperature distribution within the building envelope changes over time. A parameterized dynamic heat transfer model describes the heat transfer process of the building envelope using mathematical equations that include the thermal parameters to be identified, and reflects the dynamic changes of the heat transfer process over time. First, the material composition and physical properties of each layer of the building envelope are analyzed. Then, based on physical mechanisms such as heat conduction, convection, and radiation, the actual one-dimensional unsteady-state heat transfer process is transformed into a mathematical model characterized by parameters such as thermal resistance and heat capacity. Simultaneously, the key thermal parameters of the building envelope are incorporated as unknown parameters into the model, forming a parameterized dynamic heat transfer model. The model includes core elements such as state variables, control inputs, and observed outputs, and each element is correlated with the thermal parameters to be identified. By equivalently modeling the one-dimensional unsteady-state heat transfer process and constructing a parameterized model, the model conforms to the actual heat transfer laws of the building envelope, providing an accurate model foundation for the subsequent efficient acquisition of parameter information.

[0061] S2: Construct the Fischer information matrix based on the parameterized dynamic heat transfer model;

[0062] Specifically, the Fischer information matrix is ​​a mathematical matrix that quantifies how much information experimental data can provide for the thermal parameters to be identified. Based on the parameterized dynamic heat transfer model, the parameter sensitivity information of the model output regarding the thermal parameters to be identified is obtained, and a Fischer information matrix is ​​constructed by combining this with the experimental measurement error characteristics. This quantitatively characterizes the correlation between the output response of the building envelope under wind tunnel dynamic excitation conditions and the thermal parameters to be identified. For example, by analyzing the sensitivity of wall heat flow to thermal resistance and heat capacity in the model, and combining this with the measurement error of the temperature sensor, a corresponding Fischer information matrix is ​​constructed.

[0063] S3: Using the objective function value corresponding to the Fischer information matrix as the optimization objective, solve the wind tunnel experimental environment input signal to obtain the target dynamic excitation signal;

[0064] Specifically, the objective function value is a numerical value obtained by calculating the Fischer information matrix. The larger the objective function value, the more parameter information the experimental data can provide, and the higher the recognition accuracy. The target dynamic excitation signal is the wind tunnel excitation signal with the highest information efficiency, which can effectively separate the responses of each parameter. With the goal of maximizing the objective function value corresponding to the Fischer information matrix, the variation patterns of environmental parameters are continuously adjusted to ultimately find the optimal parameter variation sequence. By finding the target dynamic excitation signal with the highest information efficiency, the responses of each thermal parameter to be identified in the building envelope are stimulated.

[0065] S4: Conduct experimental tests on the enclosure structure under the action of the target dynamic excitation signal, obtain experimental data, and update the Fischer information matrix based on the obtained experimental data;

[0066] Specifically, the experimental data are the measured surface temperature and heat flow of the enclosure structure under the action of the target dynamic excitation signal; the wind tunnel experimental environment is adjusted according to the target dynamic excitation signal, and the experimental data is collected synchronously. During the experiment, the Fischer information matrix is ​​updated with the collected experimental data at fixed time intervals.

[0067] S5: When the preset experimental termination condition is met according to the updated Fischer information matrix, the acquired experimental data is used as the target data.

[0068] Specifically, as the target dynamic excitation signal continues to act on the enclosure structure, experimental data such as surface temperature and heat flux density are continuously collected, and the Fischer information matrix is ​​updated in real time accordingly. The updated Fischer information matrix reflects the comprehensive identification information provided by the collected data for each thermal parameter to be identified up to the current moment. When the growth rate of the determinant value of the updated Fischer information matrix continues to decrease and meets the preset experimental termination condition, it indicates that the data added by continuing to extend the experiment can no longer significantly improve the parameter identification effect, and the experiment has actually entered the information saturation stage. In this case, the experimental data already acquired at this time is determined as the target data. On the one hand, this avoids the continued collection of a large amount of low-gain, highly repetitive data, reducing the experimental cycle and energy consumption costs. On the other hand, it also ensures that the subsequent joint decoupling identification uses a set of data with high information density and strong parameter discrimination ability.

[0069] S6: Based on the target data and the parameterized dynamic heat transfer model, the thermal parameters to be identified are jointly decoupled and identified to obtain the target thermal parameters.

[0070] Specifically, the state variables, input variables, and output variables of the heat transfer process within the building envelope, as well as the identified thermal parameters characterizing the dynamic thermal performance of the building envelope, are determined. For example, maximum likelihood estimation or nonlinear least squares method is used to fit the parameter vector θ. Due to the input excitation φ... opt(t) The dynamic contributions of each parameter have been actively separated physically, enabling the estimation algorithm to converge quickly and stably to the global optimum θ. hat Furthermore, an uncertainty quantification assessment was conducted. In this result, the thermal resistance R... i Heat capacity C iThe statistical correlation between the estimated values ​​of surface heat transfer parameters h0, n, and ε is significantly reduced, achieving effective mathematical decoupling. High-quality target data is used to solve for the true thermal parameters of the building envelope, addressing the problems of multiple solutions and low accuracy in parameter identification in existing technologies. Joint decoupling identification can solve for all parameters simultaneously, considering the interactions between parameters, resulting in more comprehensive and accurate identification results. Target data under optimal signal conditions allows for parameter response separation, enabling the algorithm to quickly converge to the true values ​​and avoiding multiple solutions.

[0071] In an optional embodiment, S1 includes:

[0072] S11: Perform an equivalent transformation on the one-dimensional unsteady heat transfer process to obtain a thermal network model;

[0073] Specifically, based on the one-dimensional unsteady heat transfer process of the building envelope, the building envelope is simplified and modeled from the perspective of heat transfer mechanism. The building envelope composed of multiple layers of materials is equivalent to a thermal network model including several thermal resistance elements and several thermal capacity elements. Each thermal resistance element is used to characterize the thermal resistance between adjacent layers, and each thermal capacity element is used to characterize the heat storage capacity of the corresponding material layer or equivalent layer.

[0074] S12: Perform state-space transformation on the thermal network model to obtain a standard state-space form model;

[0075] Specifically, the thermal network model is transformed into a state-space form according to the energy conservation relationship, so that the model can be used for numerical simulation and parameter sensitivity analysis. By selecting state variables, input variables, and output variables, state equations and output equations describing the dynamic heat transfer behavior of the building envelope are established. For example, the state variable x(t) represents the temperature of each RC node. For a third-order model, x(t) = [T in(t) ,T core(t) ,T out(t) ] T , representing the temperatures of the inner surface, core layer, and outer surface, respectively. Control input u (t) : Represents a vector of environmental parameters that the wind tunnel can accurately reproduce and actively control. To achieve simulation of real, complex climates, u (t) At least including: the air temperature T acting on the surface of the specimen air(t) Relative humidity (RH) air(t) Total solar irradiance G sol(t) Wind speed v wind(t) And the equivalent sky temperature T used to simulate long-wave radiation from the sky. sky(t) Observation output y (t) : Represents a vector of physical quantities that can be measured synchronously in the experiment, typically including the temperature T of the inner and outer surfaces of the specimen. s(t) and heat flux density q (t) .

[0076]

[0077] The structure of the system matrices A, B, C, and D, and all non-zero elements are uniquely determined by the physical parameter vector θ to be identified.

[0078] S13: Associate the thermal parameters to be identified with the model parameters in the state-space form model to obtain a parameterized dynamic heat transfer model.

[0079] Specifically, the parameter set may include the thermal resistance and heat capacity of each material layer or equivalent layer, dynamic characteristic parameters of the surface convective heat transfer coefficient, and long-wavelength emissivity of the outer surface, etc. A typical parameter set is defined as:

[0080]

[0081] R1, R2, R i C1, C2, and C3 represent the thermal resistances of each material layer or equivalent layer, respectively. C i These represent the heat capacity of each material layer or equivalent layer, directly determining the insulation and heat storage capacity of the component, and are the basis for calculating the thermal inertia index D and the heat storage coefficient S; h0,n are the dynamic characteristic parameters of the surface convective heat transfer coefficient; considering that convective heat transfer is closely related to wind speed, it is modeled as... The nonlinear dynamic characteristics of convective heat transfer are accurately captured by identifying the form of ε, where ε is the long-wavelength emissivity of the outer surface, a key physical property parameter for accurately calculating the radiative heat transfer between the specimen and the sky and surrounding environment. Incorporating ε into θ enables physical decoupling and independent identification of radiative and convective heat transfer.

[0082] In an optional embodiment, S2 includes:

[0083] S21: Based on the parameterized dynamic heat transfer model, calculate the partial derivatives of the model output response with respect to the thermal parameters to be identified, and obtain the parameter sensitivity function value;

[0084] Specifically, the parameter sensitivity function value is used to quantitatively describe the contribution of different identified thermal parameters to the change of the measurable physical quantity under given experimental input conditions. The larger the parameter sensitivity function value, the more significant the impact of parameter changes on the output. For example, a large sensitivity value for heat capacity means that a slight change in heat capacity will result in a significant change in the heat flux across the wall. The computational model predicts the output. The partial derivative with respect to each component of the parameter θ, i.e., the parameter sensitivity function: This invention preferably employs automatic differentiation technology for efficient and accurate calculation. The parameter sensitivity function value quantifies the degree of influence of parameters on the output, making the subsequent construction of the Fischer information matrix more accurate, and also allowing the excitation signal design to focus on stimulating the response of key parameters, thereby improving recognition efficiency.

[0085] S22: Construct the Fischer information matrix based on the parameter sensitivity function values ​​at each discrete time point and the covariance matrix of the measurement noise.

[0086] Specifically, after obtaining the parameter sensitivity function at each discrete time point, a discrete Fischer information matrix is ​​constructed by combining it with the covariance matrix of the experimental measurement noise. The Fischer information matrix is ​​obtained by accumulating the parameter sensitivity at each sampling time within the experimental time range. Its determinant is used to characterize the amount of joint information contained in the experimental data for the thermal parameters to be identified under the current experimental design conditions. The Fischer information matrix is ​​a symmetric positive definite matrix and can be directly used for subsequent operations such as inversion and determinant calculation. First, the covariance matrix of the measurement noise is determined based on the accuracy of the measuring instrument. Then, several time points in the experiment are selected. Finally, combined with the parameter sensitivity function values, each element of the matrix is ​​calculated according to a fixed formula to form the Fischer information matrix. Assuming the measurement noise is Gaussian white noise with a covariance matrix of Q, then the Fischer information matrix based on N discrete time points is a P×P symmetric positive definite matrix (P being the number of parameters), and its (j,k) elements are:

[0087]

[0088] The inverse matrix M of FIM {-1} Mathematically, it equals the Cramér-Rao lower bound of the covariance matrix of any unbiased parameter estimator. This means that the larger det(M) is, the higher the theoretically achievable accuracy of parameter estimation, and the smaller the volume of the joint confidence ellipsoid. Constructing a matrix that accurately quantifies the information content of experimental data provides a comparable quantitative indicator for subsequent optimal excitation signal design. Considering the impact of measurement errors, the matrix should characterize the effective information content of the experimental data, avoiding information evaluation bias caused by noise interference.

[0089] In an optional embodiment, S3 includes:

[0090] S31: Parameterize the continuous-time wind tunnel experimental environment input signal to obtain finite-dimensional optimization variables;

[0091] Specifically, the continuous-time wind tunnel experimental environment input signal u(t) is parameterized in the time domain and transformed into a finite-dimensional optimization variable, for example, by using piecewise constants or specific basis function combinations. Under strict physical constraints and safety limits of the wind tunnel equipment, numerical optimization algorithms such as Sequential Quadratic Programming (SQP) or Genetic Algorithm (GA) are used to solve the above nonlinear constrained optimization problem. For example, using the piecewise constant method, the time signal u(t) is divided into several time intervals, and the signal amplitude remains constant within each time interval. The time interval [0, T] is divided into N smaller intervals, and within each smaller interval, the signal u(t) is set to a constant, denoted as . These constants are the variables we want to optimize, forming a finite-dimensional optimization variable vector. By using the controllable environmental excitation in wind tunnel experiments as optimization variables, the experimental input signal is parameterized in the time domain, transforming the continuous-time dynamic excitation signal into a finite-dimensional parameter vector.

[0092] In an optional embodiment, S31 includes:

[0093] S311: Obtain the continuous-time variation curves of each environmental parameter in the wind tunnel experimental environment input signal;

[0094] Specifically, the input signal of a wind tunnel experiment is usually not a single quantity, but rather a combination of multiple environmental parameters such as air temperature, relative humidity, total solar irradiance, wind speed, and equivalent sky temperature. Therefore, it is first necessary to determine which environmental parameters should be used as excitation inputs for subsequent optimization, based on the heat transfer boundary conditions of the building envelope under test, and further obtain the continuous-time variation curves of each environmental parameter within the experimental time domain. This acquisition can be done by directly constructing an initial continuous curve based on the pre-set experimental range, or by forming a continuous expression based on historical meteorological data, standard operating condition data, or a disturbance range set by the researcher. The purpose is to first fully describe the variation trend of each environmental parameter throughout the entire experimental period, providing a foundation for subsequent unified segmentation and parameter extraction. By first establishing continuous-time variation curves, the originally dispersed environmental boundary conditions can be incorporated into the same time-domain framework for processing, giving the subsequent parameterization process a clear object and ensuring that the generated excitation signal still corresponds to the actual, applicable wind tunnel environmental changes in a physical sense.

[0095] S312: Divide each of the continuous time variation curves into segments according to a preset time length to obtain multiple time sub-intervals;

[0096] Specifically, based on the obtained continuous-time variation curves of various environmental parameters, the entire experimental time period needs to be divided according to a preset time length. This breaks down the originally continuously changing input process into multiple time sub-intervals that facilitate optimization. The preset time length can be comprehensively set based on the wind tunnel control accuracy, the thermal response inertia of the enclosure structure, and the total experimental duration. For example, for specimens with slow thermal response, the duration of a single time sub-interval can be appropriately extended, while for inputs with faster changes, such as wind speed pulses or radiation disturbances, a finer segmentation method can be adopted to preserve key dynamic characteristics. After this step, each continuous-time variation curve is mapped to a set of sequentially arranged time sub-intervals. Subsequent processing no longer directly deals with infinite-dimensional continuous signals, but instead focuses on processing parameter expressions within a finite number of intervals. This not only significantly reduces the complexity of the optimization problem but also transforms the design of wind tunnel experimental inputs from a continuous function level to a piecewise control level, which is more in line with control methods in engineering implementation.

[0097] S313: For each of the aforementioned time sub-intervals, determine the parameterized representation value of the corresponding environmental parameter within that time sub-interval;

[0098] Specifically, for each time sub-interval, it is necessary to further determine the parameterized representation of the corresponding environmental parameter within that interval. This is essentially extracting a parameter representation that can represent the changing state of each continuous input segment. If a piecewise constant form is used, the parameterized representation can be the amplitude that remains unchanged within the time sub-interval; if a piecewise linear form is used, it can be characterized by the starting value, ending value, or slope of the interval; if a basis function expansion form is used, the parameterized representation can also be the coefficients of the corresponding basis function. In this way, each time sub-interval is no longer an abstract continuous process, but is transformed into a parameter quantity that can participate in calculations and comparisons. The purpose of this step is to preserve the time-domain variation characteristics of the environmental parameters while compressing them into a finite parameter representation that is easy to solve, so that the strength, direction, and duration of disturbances of different environmental parameters in different time periods can be clearly expressed, thereby providing a direct basis for the subsequent construction of candidate dynamic excitation signals.

[0099] S314: Generate a discrete parameter sequence corresponding to each environmental parameter based on the parameterized representation value corresponding to each time sub-interval;

[0100] Specifically, after the parameterized representation values ​​of each time sub-interval are determined, the parameterized representation values ​​of the same environmental parameter within each time sub-interval need to be arranged sequentially according to time order to generate a discrete parameter sequence corresponding to that environmental parameter. Taking air temperature as an example, if the entire experimental period is divided into several sub-intervals, each sub-interval corresponds to a temperature parameter value. These parameter values, arranged in chronological order, form a discrete temperature parameter sequence. Other environmental parameters such as wind speed and radiation are processed in the same way. The resulting discrete parameter sequence is actually a compressed expression of continuous-time input in a discrete-time structure. It not only preserves the chronological relationship in the time dimension but also retains the information on the changes in input intensity within each sub-interval. By generating a discrete parameter sequence, the organization of subsequent optimization variables can be transformed from single-point parameters to sequential parameters, enabling the optimization algorithm to directly solve for the temporal changes of each environmental parameter throughout the entire experimental process, thereby improving the overall integrity and specificity of the target dynamic excitation signal design.

[0101] S315: Combine the discrete parameter sequences corresponding to each environmental parameter to obtain finite-dimensional optimized variables used to characterize the input signal of the wind tunnel experimental environment.

[0102] Specifically, after obtaining the discrete parameter sequences corresponding to each environmental parameter, these sequences need to be further combined according to a unified time-domain structure to form finite-dimensional optimization variables that can comprehensively characterize the wind tunnel experimental environment input signal. This combination is not a simple splicing, but rather incorporates the parameter values ​​of multiple environmental parameters, such as air temperature, relative humidity, total solar irradiance, wind speed, and equivalent sky temperature, across various time sub-intervals into the same set of optimization variables. This ensures that each set of finite-dimensional variables corresponds to a complete composite environmental input scheme. The resulting finite-dimensional optimization variables retain the coupling characteristics of the combined effects of multiple environmental parameters while transforming the original continuous-time control problem into a finite-dimensional solution problem that can be directly handled by numerical optimization algorithms. The wind tunnel experimental environment input signal is ultimately transcribed into a set of variables with a clear structure, defined dimensions, and ease of constraint application and optimization execution.

[0103] S32: Based on the finite-dimensional optimization variables, construct the corresponding time series function and generate candidate dynamic excitation signals;

[0104] Specifically, candidate dynamic excitation signals are intermediate excitation signals generated during the optimization algorithm iteration process based on a set of values ​​for finite-dimensional optimization variables. Based on these finite-dimensional optimization variables, the parameters are sequentially concatenated or superimposed according to time division to construct a continuous time series function that changes with time, thus obtaining a complete dynamic input curve, such as a sequence of air temperature, solar radiation, or wind speed changing with time. Furthermore, under different combinations of optimization variables, multiple different time series functions can be generated, each corresponding to a set of candidate dynamic excitation signals. These candidate signals are used as inputs to the parameterized dynamic heat transfer model for simulation analysis and Fischer information matrix calculation to select the optimal dynamic excitation scheme.

[0105] S33: Based on the preset parameter nominal values, the candidate dynamic excitation signal is input into the parameterized dynamic heat transfer model to determine the candidate output response signal;

[0106] Specifically, the D-optimal experimental design criterion is adopted, which directly uses maximizing the FIM determinant as the optimization objective to solve for the optimal experimental input:

[0107]

[0108] Where, θ nom These are nominal parameter values ​​set based on prior knowledge such as material information; that is, preset nominal parameter values. Candidate excitation signals are substituted into a parameterized model with nominal values, and the model's output data is calculated. This provides specific model output data for subsequent calculations of parameter sensitivity and the construction of the Fischer information matrix, giving the information efficiency assessment of candidate excitation signals a practical basis. The nominal values ​​closely approximate the true parameter values, ensuring that the calculated candidate output response signals are more realistic, leading to more accurate subsequent information efficiency assessments.

[0109] S34: Calculate the objective function value of each Fischer information matrix based on each candidate output response signal;

[0110] Specifically, using candidate excitation signals as the basis for experimental design and candidate output response signals as the foundation for model output, the method calculates parameter sensitivity, constructs a Fischer information matrix, and finally calculates the determinant value of the matrix. This transforms the information efficiency of candidate excitation signals into concrete numerical values, allowing direct comparison of the information efficiency of different candidate signals and providing a basis for algorithm iteration to find the optimal signal. Beneficial effects: A unified calculation method ensures consistent evaluation standards for different candidate signals, making comparisons more objective; the determinant value calculation results accurately quantify information efficiency, making the direction of algorithm iteration clearer.

[0111] S35: Compare the objective function values ​​and take the candidate dynamic excitation signal corresponding to the Fischer information matrix with the largest objective function value as the target dynamic excitation signal.

[0112] Specifically, each objective function value is recorded in real time, and the values ​​are compared to find the candidate dynamic excitation signal corresponding to the Fischer information matrix with the largest objective function value. Maximizing the Fischer information matrix is ​​used as the objective function for the experimental design, constructing a nonlinear optimization problem with physical constraints, and solving the problem using a numerical optimization algorithm. By solving this optimization problem, a set of dynamic input parameters that maximize the amount of parameter identification information under given experimental duration and equipment constraints is obtained, thereby generating a target dynamic excitation signal adapted to the wind tunnel experiment. The obtained target dynamic excitation signal φ opt(t) It is a dynamic climate sequence with extremely high information density. Its design contains a clear intention to decouple: the sequence includes large, non-periodic temperature and solar radiation steps, aiming to strongly excite the heat capacity C. i transient response and thermal resistance R i The steady-state effect is observed; simultaneously, the sequence integrates wind speed pulses with varying amplitudes across multiple time scales and sky temperature fluctuations, specifically designed to differentiate and separate the wind speed-sensitive convective heat transfer coefficient (h0,n) and the dynamic response related to the radiation characteristic ε. Therefore, the required optimal total experimental duration T... opt This method can shorten the testing cycle by more than 50% compared to traditional methods that reproduce natural weather conditions over several days. The target signal has optimal information efficiency, allowing subsequent experiments to collect high-quality data in the shortest possible time. Equipment constraint checks ensure that the signal can be actually executed, avoiding the problem of theoretically optimal results that cannot be achieved.

[0113] Please see Figure 2 , Figure 2 This is an example diagram of a set of optimal target dynamic excitation signals obtained based on the Fischer information matrix optimality criterion. This set of signals defines the time-varying characteristics of five environmental parameters that need to be simultaneously applied to the enclosure structure specimen in a wind tunnel experiment. Its core design objective is to actively separate the coupling contributions of parameters such as thermal resistance, heat capacity, convective heat transfer coefficient, and radiation characteristics to the thermal flux response from the excitation source. Among them:

[0114] Wind speed: Combining low-frequency slow variation and high-frequency pulse, two parameters of the convective heat transfer coefficient are separated;

[0115] Temperature: Large, non-periodic step changes, distinguishing the response time between thermal resistance and thermal capacity;

[0116] Solar radiation: Rectangular pulse sequence, independently exciting the radiation characteristics of the outer surface;

[0117] Effective sky temperature: changes in tandem with temperature, enhancing orthogonal identification of radiation parameters;

[0118] Relative humidity: changes periodically in the opposite phase to temperature, providing data support for wet operating conditions.

[0119] from Figure 2 As shown in the waveforms, the wind speed curve uses a combination of gradually varying segments and short-duration pulse segments, ensuring that the surface convection heat transfer process is fully excited at different time scales, which is beneficial for enhancing the identification sensitivity of wind speed-related heat transfer parameters. The air temperature curve uses multiple segments of non-periodic step changes, superimposed with continuous fluctuations of a certain amplitude, to differentiate the transient heat storage response and steady-state heat transfer response of the building envelope in the time domain, so as to better distinguish the effects of heat capacity parameters and thermal resistance parameters. The total solar radiation irradiance curve uses multiple sets of rectangular pulses to enhance the thermal response changes of the outer surface after radiation excitation, and improve the identification capability of relevant parameters under radiation boundary conditions. The sky equivalent temperature curve and the air temperature curve maintain a related but not completely synchronized relationship to simulate the independent disturbance of the long-wave radiation heat transfer boundary, enhancing the identifiability of the radiation characteristics of the outer surface. The air relative humidity curve uses a periodic fluctuation form and forms an interleaved relationship with other environmental parameters, making the overall experimental boundary conditions closer to the actual dynamic service environment, and providing a basis for thermal testing and extended analysis under complex environmental coupling conditions.

[0120] In an optional embodiment, S4 includes:

[0121] S41: Under the action of the target dynamic excitation signal, the enclosure structure is tested experimentally, and corresponding experimental data are collected at multiple discrete time points;

[0122] S42: Based on the experimental data at each discrete time point and the parameterized dynamic heat transfer model, determine the sensitivity function value of the target parameter of the thermal parameter to be identified in the model output response;

[0123] In an optional embodiment, S42 includes:

[0124] S421: Based on the experimental data corresponding to each discrete time point, extract the target observation output of the enclosure structure at each discrete time point;

[0125] Specifically, the process begins by organizing and extracting the raw experimental data collected at each discrete time point during the experiment, based on the observation types upon which the Fischer information matrix update depends. This results in target observation outputs that directly characterize the instantaneous thermal response state of the building envelope. The experimental data typically includes internal and external surface temperatures, heat flux density, and input information corresponding to the boundary environment. However, not all raw sampled values ​​are directly used in subsequent sensitivity calculations. Instead, observation results corresponding to the model output variables are extracted to ensure consistency between the experimental data and the model output. This process ensures that each discrete time point corresponds to a clear set of target observation outputs, reflecting the actual heat transfer performance of the building envelope under the target dynamic excitation signal at that moment. It also provides a unified data entry point for subsequent state matching, avoiding computational biases caused by the chaotic dimensions and mixed types of the raw data, and improving the relevance of subsequent parameter sensitivity analysis.

[0126] S422: Match the target observation output corresponding to each discrete time point with the parameterized dynamic heat transfer model to determine the model state estimation result corresponding to each discrete time point;

[0127] Specifically, matching the target observation output at each discrete time point with the parameterized dynamic heat transfer model essentially establishes a correspondence between the experimentally measured real response and the theoretical state within the model, ensuring that the model at each discrete time point falls into a position as consistent as possible with the actual thermal state of the specimen. In practice, based on the target dynamic excitation signal, observation output, and the state-space expression of the parameterized dynamic heat transfer model at that discrete time point, the state estimation results of the internal heat capacity nodes, surface nodes, or equivalent heat transfer nodes of the enclosure structure at that moment can be deduced. This prevents the model from remaining at the level of purely theoretical prediction but rather updates synchronously with the actual thermal state during the experiment. Through this state matching process, the parameter sensitivity calculated subsequently is not merely a theoretical partial derivative under some ideal operating condition but is based on the local response characteristics constrained by experimental data, thereby enhancing the realism of the sensitivity assessment and its adaptability to the current experimental stage.

[0128] S423: Based on the model state estimation results corresponding to each discrete time point, calculate the local parameter sensitivity of the model output response to each thermal parameter to be identified;

[0129] Specifically, after obtaining the model state estimation results for each discrete time point, it is necessary to further calculate the local parameter sensitivity of the model output response to each identified thermal parameter based on these state estimation results. This local parameter sensitivity can be understood as the extent to which the model output response changes when a specific thermal parameter undergoes a small change near the actual thermal state at the current discrete time point. Since the thermal resistance, heat capacity, and surface heat transfer parameters of the building envelope have different degrees of influence on the output under different thermal states, this sensitivity calculation method based on local states can more accurately characterize the identification contribution of each parameter in the current experimental stage. The result obtained is no longer a globally average parameter influence, but rather locally changing information over time. This is beneficial for identifying which parameters are fully excited at the current moment and which parameters are still in a weak response state, thus laying the foundation for subsequent screening of high-information moments and updating the Fischer information matrix.

[0130] S424: Based on the change range of local parameter sensitivity at each discrete time point, select the effective discrete time points that meet the preset sensitivity conditions.

[0131] Specifically, after obtaining the local parameter sensitivity at each discrete time point, it is necessary to filter out the effective discrete time points that meet the preset sensitivity conditions based on the magnitude of the changes in these local parameter sensitivities. This step is added because not all sampling moments have equal value for parameter identification. At some moments, although data acquisition has been completed, the corresponding local parameter sensitivity is low due to the relatively gradual change in the enclosure structure response, the insignificant parameter influence, or the strong coupling between different parameters. Including these moments in subsequent calculations with equal weight might dilute the role of high-information data. By comparing the magnitude of changes in local parameter sensitivity at each discrete time point and filtering according to the preset sensitivity conditions, moments that contribute more significantly to parameter identification and have a stronger ability to distinguish different parameters can be retained, while low-contribution or weakly sensitive moments can be eliminated. This ensures that the subsequently constructed target parameter sensitivity function value more comprehensively reflects the effective experimental information.

[0132] S425: Generate the target parameter sensitivity function value corresponding to each thermal parameter to be identified based on the local parameter sensitivity corresponding to each effective discrete time point.

[0133] Specifically, after selecting the effective discrete time points, it is necessary to further generate target parameter sensitivity function values ​​for each thermal parameter to be identified based on the local parameter sensitivities corresponding to these effective discrete time points. This generation process is not a simple reference to the results of a single moment, but rather the organization and aggregation of local parameter sensitivities from multiple effective time points in chronological order. This ensures that each thermal parameter to be identified forms a target parameter sensitivity function that reflects its actual identifiability at the current experimental stage. The resulting target parameter sensitivity function values ​​retain dynamic changes over time while avoiding interference from inefficient data, making them more suitable for incremental updates of the Fischer information matrix. Through this step, the originally discrete local sensitivities are transformed into functional results that directly support information content evaluation, creating a closed-loop connection between experimental data, state estimation, and parameter identification.

[0134] S43: Update the Fischer information matrix based on the target parameter sensitivity function values ​​and the covariance matrix of the measurement noise at each discrete time point.

[0135] Specifically, the target dynamic excitation signal φ opt(t) Essentially, it describes the time-varying sequence of environmental parameters such as air temperature, solar radiation, wind speed, humidity, and equivalent sky temperature, requiring the dynamic excitation signal φ of the target. opt(t) The control system, loaded into the wind tunnel experimental environment, coordinates and controls each environmental adjustment unit, enabling the wind tunnel to accurately reproduce the corresponding dynamic climate conditions according to the time series. Experimental tests are conducted on the building envelope, and experimental data such as surface temperature and heat flux density are collected at multiple discrete time points. This step yields the true dynamic response of the building envelope under high information density excitation.

[0136] Then, the experimental data at each discrete time point are combined with the parameterized dynamic heat transfer model. By using the functional relationship between the input, output and the thermal parameters to be identified established by the model, the parameter sensitivity function value of the model output response to each thermal parameter to be identified is determined. In other words, it is determined how much a small change in each parameter will affect the output under the current experimental conditions. Thus, the original experimental data is further transformed into sensitivity information that can reflect the degree of parameter identifiability.

[0137] The Fischer information matrix is ​​updated based on the sensitivity function values ​​of the target parameters and the covariance matrix of the measurement noise at each discrete time point. This is equivalent to gradually accumulating the effective information brought by the newly added experimental data at the current moment to the overall information matrix, so that the Fischer information matrix can dynamically represent the comprehensive information of all the data collected up to the current moment.

[0138] In an optional embodiment, S5 includes:

[0139] S51: Based on the updated Fischer information matrix, obtain the determinant value of the Fischer information matrix at each moment within the preset time interval;

[0140] S52: Calculate the determinant growth rate based on the determinant values ​​of each Fischer information matrix;

[0141] S53: When the growth rate of the determinant is less than the preset growth rate threshold, it is determined that the preset experiment termination condition is met, and the acquired experimental data is used as the target data.

[0142] Specifically, the determinant growth rate can be determined by the relative change in the determinant value of the Fischer information matrix within adjacent time intervals. This represents the degree of contribution of newly added experimental data to the information provided for parameter identification. The smaller the determinant growth rate, the more limited the contribution of the newly added data to improving parameter identifiability. A preset growth rate threshold, such as 1%, is obtained. The growth rate is calculated in real time during the experiment. When the growth rate is less than 1% for one consecutive hour, the experimental information is considered saturated, and the experiment is terminated. All collected experimental data is then preprocessed and determined as the target data. Termination occurs promptly when sufficient experimental information is available, avoiding the waste of time and energy caused by continuing the experiment, while ensuring that the collected target data meets the requirements for high-precision parameter identification. Beneficial effects: The requirement for continuous time avoids erroneous termination due to instantaneous interference, ensuring the accuracy of the judgment; timely termination of the experiment significantly saves experimental resources; the target data is free of redundancy, and subsequent processing is more efficient.

[0143] In an optional embodiment, S6 includes:

[0144] S61: Associate the target data with the parameterized dynamic heat transfer model to determine the model output response of the parameterized dynamic heat transfer model when the thermal parameter to be identified takes different parameter values;

[0145] S62: Based on the deviation between the model output response and the target data, construct a parameter optimization problem;

[0146] S63: The parameter optimization problem is solved jointly by using the maximum likelihood estimation method or the nonlinear least squares method to obtain the target thermal parameters.

[0147] Specifically, after obtaining the target data, it is associated with a parameterized dynamic heat transfer model. By changing the values ​​of the thermal parameters to be identified, the corresponding model output response is obtained, and a parameter optimization problem is constructed based on the deviation between the model output response and the target data.

[0148] Based on this, maximum likelihood estimation or nonlinear least squares method is used to fit the parameter vector θ. This is because the input target dynamic excitation signal φ... opt(t) The dynamic contributions of each parameter have been actively separated physically, enabling the estimation algorithm to converge quickly and stably to the global optimum θ. hat In this result, the thermal resistance R i Heat capacity C i The statistical correlation between the estimated values ​​of surface heat transfer parameters h0,n,ε is significantly reduced, achieving effective parameter decoupling and identification.

[0149] Obtain parameter estimates θ hat Then, calculate the Fisher information matrix M(θ) at that point. hat Based on parameter estimation theory, the covariance matrix of the parameter estimate is obtained by approximating it using its inverse matrix, thereby determining the estimated standard deviation and confidence interval of each parameter to be identified. The standard deviation and confidence interval are compared with preset accuracy requirements. When the accuracy requirements are met, the parameter estimation result is determined as the target thermal parameter. When they are not met, the target data or model is corrected and the parameters are refitted until the accuracy requirements are met. For example, a preset accuracy requirement is obtained (e.g., standard deviation ≤ 0.03, confidence interval width ≤ 0.1). The standard deviation and confidence interval of each thermal parameter are compared with the preset requirement. If they are met, the fitted value is directly determined as the target thermal parameter; if they are not met, the target data or model is checked, corrected, and refitted until the accuracy requirements are met.

[0150] Based on the thermal resistance R identified by decoupling i and heat capacity C i This allows for further calculation of relevant indicators in engineering applications, such as the thermal inertia index D.

[0151]

[0152] Among them, S i is the heat storage coefficient of the i-th layer; n is the total number of layers.

[0153] Heat storage coefficient S i :

[0154] in, Let d be the thermal conductivity of the i-th layer. i Where T is the layer thickness and T is the design cycle, which is usually 24 hours.

[0155] Instantaneous heat storage per unit area:

[0156] Where m is the total number of heat capacity nodes, T j(t) represents the temperature of the j-th thermal capacity node at time t, which is obtained from the state estimation of the identification model.

[0157] Cumulative heat storage per unit area:

[0158]

[0159] This represents the total heat absorbed or released per unit area of ​​the building envelope from time t1 to t2.

[0160] Example 2

[0161] Please see Figure 3 This invention provides a device for decoupling and identifying thermal parameters of building envelope, the device comprising:

[0162] The model building module is used to perform equivalent modeling of the one-dimensional unsteady heat transfer process of the building envelope under test, and obtain a parameterized dynamic heat transfer model.

[0163] The matrix construction module is used to construct a Fischer information matrix based on the parameterized dynamic heat transfer model.

[0164] The excitation signal determination module is used to take the objective function value corresponding to the Fischer information matrix as the optimization objective, solve the wind tunnel experimental environment input signal, and obtain the target dynamic excitation signal.

[0165] The matrix update module is used to conduct experimental tests on the enclosure structure under the action of the target dynamic excitation signal, acquire experimental data, and update the Fischer information matrix based on the acquired experimental data.

[0166] The target data determination module is used to take the acquired experimental data as target data when it is determined that the preset experimental termination condition is met according to the updated Fischer information matrix.

[0167] The decoupling module is used to perform joint decoupling identification of the thermal parameters to be identified based on the target data and the parameterized dynamic heat transfer model, so as to obtain the target thermal parameters.

[0168] It should be noted that each module and unit in the thermal parameter decoupling identification device of the building envelope in this embodiment corresponds one-to-one with each step in the thermal parameter decoupling identification method of the building envelope in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned thermal parameter decoupling identification method of the building envelope, and will not be repeated here.

[0169] Example 3

[0170] Furthermore, the method for decoupling and identifying thermal parameters of the building envelope in this embodiment of the invention can be implemented by electronic devices. Figure 4A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention is shown.

[0171] Electronic devices may include processors and memory storing computer program instructions.

[0172] Specifically, the processor may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement embodiments of the present invention.

[0173] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0174] Computer-readable media include both permanent and non-permanent, removable and non-removable media, which can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated communication signals and carrier waves.

[0175] The processor reads and executes computer program instructions stored in the memory to implement any of the thermal parameter decoupling identification methods for the building envelope in the above embodiments.

[0176] In one example, the electronic device may also include a communication interface and a bus. For example, Figure 4 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0177] The communication interface is mainly used to enable communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0178] A bus, including hardware, software, or both, couples components of an electronic device together. For example, and not limitingly, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, a bus may include one or more buses. While specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0179] Example 4

[0180] Furthermore, in conjunction with the decoupling identification method for thermal parameters of the building envelope in the above embodiments, this invention can be implemented using a computer-readable storage medium. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the decoupling identification methods for thermal parameters of the building envelope in the above embodiments.

[0181] In summary, the decoupling identification method, apparatus, equipment, and medium for thermal parameters of building envelopes provided in this invention maximize the identifiability of thermal parameters from the source. The construction of the parameterized dynamic heat transfer model closely matches the actual dynamic heat transfer law of the building envelope. The adaptive experiment termination mechanism based on the growth of the Fischer information matrix value can accurately determine the information saturation state of the experiment, avoiding the waste of experimental resources. The joint decoupling identification combined with the parameterized dynamic heat transfer model allows the estimation uncertainty of thermal parameters to reach the mathematical minimum lower bound, significantly improving the accuracy and efficiency of key thermal parameter identification. The identification process can effectively decouple each parameter, and the obtained high-precision thermal parameters can provide more reliable basic data support.

[0182] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the accompanying drawings. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0183] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0184] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0185] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0186] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0187] It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or apparatus. However, this invention is not limited to the order of the steps described above; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0188] The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A method for decoupling and identifying thermal parameters of building envelope, characterized in that, The method includes: An equivalent model of the one-dimensional unsteady heat transfer process of the building envelope under test is performed to obtain a parameterized dynamic heat transfer model. Based on the parameterized dynamic heat transfer model, a Fischer information matrix is ​​constructed; Using the objective function value corresponding to the Fischer information matrix as the optimization objective, the wind tunnel experimental environment input signal is solved to obtain the target dynamic excitation signal; Experimental tests were conducted on the enclosure structure under the action of the target dynamic excitation signal to obtain experimental data, and the Fischer information matrix was updated based on the obtained experimental data. When the preset experimental termination condition is met according to the updated Fischer information matrix, the acquired experimental data is used as the target data. Based on the target data and the parameterized dynamic heat transfer model, the thermal parameters to be identified are jointly decoupled and identified to obtain the target thermal parameters. The step of using the objective function value corresponding to the Fischer information matrix as the optimization objective to solve for the wind tunnel experimental environment input signal to obtain the target dynamic excitation signal includes: The wind tunnel experimental environment input signal over continuous time is subjected to time-domain parameterization to obtain finite-dimensional optimization variables; Based on the finite-dimensional optimization variables, a corresponding time series function is constructed, and candidate dynamic excitation signals are generated; Based on the preset parameter nominal values, the candidate dynamic excitation signal is input into the parameterized dynamic heat transfer model to determine the candidate output response signal; Based on each candidate output response signal, the objective function value of each Fischer information matrix is ​​calculated; Compare the objective function values ​​and select the candidate dynamic excitation signal corresponding to the Fischer information matrix with the largest objective function value as the target dynamic excitation signal.

2. The method for decoupling and identifying thermal parameters of building envelope according to claim 1, characterized in that, The one-dimensional unsteady-state heat transfer process of the building envelope under test is equivalently modeled to obtain a parameterized dynamic heat transfer model, including: The one-dimensional unsteady heat transfer process is transformed into an equivalent model to obtain a thermal network model; The thermal network model is transformed into a state-space form to obtain a standard state-space model. The thermal parameters to be identified are associated with the model parameters in the state-space form model to obtain a parameterized dynamic heat transfer model.

3. The method for decoupling and identifying thermal parameters of building envelope according to claim 1, characterized in that, The construction of the Fischer information matrix based on the parameterized dynamic heat transfer model includes: Based on the parameterized dynamic heat transfer model, the partial derivatives of the model output response with respect to the thermal parameters to be identified are calculated to obtain the parameter sensitivity function value. Based on the parameter sensitivity function values ​​at each discrete time point and the covariance matrix of the measurement noise, a Fischer information matrix is ​​constructed.

4. The method for decoupling and identifying thermal parameters of building envelope according to claim 1, characterized in that, The step of conducting experimental tests on the enclosure structure under the action of the target dynamic excitation signal, acquiring experimental data, and updating the Fischer information matrix based on the acquired experimental data includes: Under the action of the target dynamic excitation signal, the enclosure structure is tested experimentally, and corresponding experimental data are collected at multiple discrete time points; Based on the experimental data at each discrete time point and the parameterized dynamic heat transfer model, the sensitivity function value of the target parameter of the thermal parameter to be identified in the model output response is determined. The Fischer information matrix is ​​updated based on the target parameter sensitivity function values ​​and the covariance matrix of the measurement noise at each discrete time point.

5. The method for decoupling and identifying thermal parameters of building envelope according to claim 1, characterized in that, When the preset experiment termination condition is met based on the updated Fischer information matrix, the acquired experimental data is used as the target data, including: Based on the updated Fischer information matrix, obtain the determinant value of the Fischer information matrix at each moment within the preset time interval; The growth rate of the determinant is calculated based on the determinant values ​​of each Fischer information matrix. When the growth rate of the determinant is less than the preset growth rate threshold, it is determined that the preset experiment termination condition is met, and the acquired experimental data is used as the target data.

6. The method for decoupling and identifying thermal parameters of building envelope according to any one of claims 1-5, characterized in that, The step of jointly decoupling and identifying the thermal parameters to be identified based on the target data and the parameterized dynamic heat transfer model to obtain the target thermal parameters includes: The target data is associated with the parameterized dynamic heat transfer model to determine the model output response of the parameterized dynamic heat transfer model when the thermal parameter to be identified takes different values. Based on the deviation between the model output response and the target data, a parameter optimization problem is constructed. The target thermal parameters are obtained by jointly solving the parameter optimization problem using either the maximum likelihood estimation method or the nonlinear least squares method.

7. A device for decoupling and identifying thermal parameters of an enclosure structure, characterized in that, The device includes: The model building module is used to perform equivalent modeling of the one-dimensional unsteady heat transfer process of the building envelope under test, and obtain a parameterized dynamic heat transfer model. The matrix construction module is used to construct a Fischer information matrix based on the parameterized dynamic heat transfer model. The excitation signal determination module is used to take the objective function value corresponding to the Fischer information matrix as the optimization objective, solve the wind tunnel experimental environment input signal, and obtain the target dynamic excitation signal. The matrix update module is used to conduct experimental tests on the enclosure structure under the action of the target dynamic excitation signal, acquire experimental data, and update the Fischer information matrix based on the acquired experimental data. The target data determination module is used to take the acquired experimental data as target data when it is determined that the preset experimental termination condition is met according to the updated Fischer information matrix. The decoupling module is used to perform joint decoupling identification of the thermal parameters to be identified based on the target data and the parameterized dynamic heat transfer model, so as to obtain the target thermal parameters. The step of using the objective function value corresponding to the Fischer information matrix as the optimization objective to solve for the wind tunnel experimental environment input signal to obtain the target dynamic excitation signal includes: The wind tunnel experimental environment input signal over continuous time is subjected to time-domain parameterization to obtain finite-dimensional optimization variables; Based on the finite-dimensional optimization variables, a corresponding time series function is constructed, and candidate dynamic excitation signals are generated; Based on the preset parameter nominal values, the candidate dynamic excitation signal is input into the parameterized dynamic heat transfer model to determine the candidate output response signal; Based on each candidate output response signal, the objective function value of each Fischer information matrix is ​​calculated; Compare the objective function values ​​and select the candidate dynamic excitation signal corresponding to the Fischer information matrix with the largest objective function value as the target dynamic excitation signal.

8. An electronic device, characterized in that, include: The method comprises at least one processor, at least one memory, and computer program instructions stored in the memory, wherein when the computer program instructions are executed by the processor, the method for decoupling and identifying thermal parameters of the building envelope as described in any one of claims 1-6 is implemented.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, the method for decoupling and identifying thermal parameters of the building envelope as described in any one of claims 1-6 is implemented.

Images