Thermoelectric power generation module temperature estimation method based on improved thermal network model
By introducing nonlinear heat dissipation branch and NSGA-II algorithm to extract thermal parameters in the equivalent thermal network model, the accuracy of temperature difference estimation of the hot and cold ends of the temperature difference generator sheet is solved, and fast and accurate temperature difference estimation is achieved, and the thermal power is improved. Performance evaluation and prediction accuracy of power generation systems.
Patent Information
- Application Number
- CN202411193325.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-03
AI Technical Summary
When estimating the temperature difference between the hot and cold ends of the temperature difference power generator, the existing equivalent thermal network model cannot accurately describe the nonlinear heat dissipation characteristics, resulting in large estimation errors and difficult to meet engineering needs.
Nonlinear convective heat dissipation and radiative heat dissipation branches are introduced in the traditional equivalent thermal network model. By improving the equivalent thermal network model and extracting thermal parameters in combination with the NSGA-II algorithm, a multi-objective function is constructed to improve the accuracy of temperature difference estimation.
It realizes rapid and accurate estimation of the temperature difference between the hot and cold ends of the temperature difference power generation sheet, reduces the estimation error, and improves the performance evaluation and prediction accuracy of the thermoelectric power generation system.
Smart Images

Figure CN119066973A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of temperature difference energy power generation, and in particular relates to a temperature estimation method for a thermoelectric power generation module based on an improved thermal network model. Background Art
[0002] Thermoelectric generator (TEG) is a semiconductor device that collects heat and converts it into electrical energy based on the Seebeck effect of thermoelectric materials. Due to its simple structure, easy use and environmental friendliness, it has been applied in many fields such as aerospace, transportation, and energy. In particular, in the field of low-power wireless monitoring system energy supply, thermoelectric generators have shown great commercial value. The output performance of thermoelectric generators is not only related to the physical properties of the material, but also to the temperature difference between the hot and cold ends of the thermoelectric generator. When conducting performance evaluation and prediction, it is necessary to rely on thermoelectric models to estimate the temperature difference between the hot and cold ends of the thermoelectric generator because real-time temperature measurement data cannot be obtained. Thermoelectric models usually consist of two parts: an electrical model and a thermal model. Among them, the electrical model describes the changes in the output voltage and power generation of the thermoelectric generator under different hot and cold end temperature differences, while the thermal model describes the thermal characteristics of the internal materials of the thermoelectric generator and the heat dissipation characteristics to the external environment. Therefore, accurately establishing a thermal model is the key to estimating the temperature difference between the hot and cold ends, and is also an important guarantee for the performance evaluation and prediction accuracy of the thermoelectric system.
[0003] The equivalent thermal network model method is one of the methods for establishing the thermal model of the thermoelectric generator. The equivalent thermal network model method simplifies the complex thermal analysis into an electrical calculation problem through analogy. Although the current equivalent thermal network model has played an important role in the thermoelectric modeling of the thermoelectric generator, it is still a great challenge to quickly and accurately estimate the temperature difference between the hot and cold ends of the thermoelectric generator in engineering. According to the basic principles of heat transfer, objects usually release heat to the surrounding environment through thermal convection and thermal radiation. Since temperature affects the flow of air in the boundary layer of the object surface, there is a nonlinear relationship between convective heat dissipation and the object temperature, and this nonlinear phenomenon is particularly obvious under natural convection conditions. The radiation heat exchange between the object and the environment is related to the fourth power of the object temperature, so it also presents nonlinear characteristics. Under the nonlinear effects of thermal convection and thermal radiation, the heat dissipation power of the thermoelectric generator is not only nonlinear, but also time-varying. However, due to complex factors such as material, structure and surface technology, the nonlinear heat dissipation characteristics of the thermoelectric generator are difficult to calculate theoretically in engineering. This results in most equivalent thermal network models ignoring the nonlinear effects of thermal convection and thermal radiation on the temperature of the thermoelectric generator, and failing to accurately describe the temperature difference between the cold and hot ends. Summary of the invention
[0004] The purpose of the present invention is to propose a temperature estimation method for a thermoelectric power generation module based on an improved thermal network model, so as to overcome the insufficiency that the nonlinear heat dissipation characteristics of the thermoelectric power generation sheet are difficult to accurately calculate, and effectively obtain the temperature difference between the cold and hot ends and its dynamic changes.
[0005] The technical solution to achieve the purpose of the present invention is: a method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model, the steps of the estimation method are as follows:
[0006] Step 1: According to the nonlinear principle of radiation heat dissipation and convection heat dissipation, based on the traditional equivalent thermal network model, nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator are introduced into the traditional external thermal network respectively, and the equivalent thermal network model is designed and improved, and then go to step 2.
[0007] Step 2: Construct a multi-objective function based on the improved equivalent thermal network model, and extract the thermal parameters of the thermoelectric generator by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results, and then proceed to step 3.
[0008] Step 3: Using the improved equivalent thermal network model and identification results, under given working conditions, estimate the temperature of the thermoelectric generator to obtain the estimated temperature difference between the cold and hot ends.
[0009] Compared with the prior art, the present invention has the following significant advantages:
[0010] (1) With the advantages of simple thermal network principle and small computational complexity, this method can quickly estimate the temperature difference between the hot and cold ends under specific conditions.
[0011] (2) If the future working conditions of the thermoelectric power generation unit can be known in advance, the temperature difference between the hot and cold ends can be estimated predictively, providing an important basis for the performance prediction and reliability analysis of the entire thermoelectric power generation system. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 It is a flow chart of the present invention.
[0013] Figure 2 Schematic diagram of the basic structure of a traditional equivalent thermal network according to an embodiment of the present invention.
[0014] Figure 3 Schematic diagram of the basic structure of the improved equivalent thermal network according to an embodiment of the present invention.
[0015] Figure 4 is the heat source power in the parameter identification experiment of the embodiment of the present invention.
[0016] Figure 5 Parameter identification results in the improved equivalent thermal network according to an embodiment of the present invention.
[0017] Figure 6 The heat source power and the ambient temperature in the cold-hot end temperature difference estimation experiment of the embodiment of the present invention.
[0018] Figure 7 The hot end temperature estimation of the thermoelectric power generation sheet according to the embodiment of the present invention is compared with the actual measurement result.
[0019] Figure 8 The cold-end temperature estimation of the thermoelectric generator according to the embodiment of the present invention is compared with the actual measurement result.
[0020] Fig. 9 The temperature estimation of the heat sink of the thermoelectric power generation sheet according to the embodiment of the present invention is compared with the actual measurement result.
[0021] Fig.10 ∈ R ∈ R , which is the temperature estimation error of the traditional equivalent thermal network model in the embodiment of the present invention.
[0022] Fig.11 It is the temperature estimation error of the improved equivalent thermal network model according to the embodiment of the present invention. DETAILED DESCRIPTION
[0023] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.
[0024] The present invention proposes a temperature estimation method for a thermoelectric power generation module based on an improved thermal network model. The traditional equivalent thermal network model is improved by introducing nonlinear convection heat dissipation and radiation heat dissipation branches, and multi-objective identification is used to extract thermal parameters in the thermal network offline, thereby realizing rapid estimation of the temperature difference between the cold and hot ends of a thermoelectric power generation piece.
[0025] like Figure 1 As shown, a method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model comprises:
[0026] Step 1: According to the nonlinear principle of radiation heat dissipation and convection heat dissipation, based on the traditional equivalent thermal network model, nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator are introduced into the traditional external thermal network respectively to design an improved equivalent thermal network model.
[0027] The traditional equivalent thermal network model in step 1 is as follows Figure 2 As shown, its mathematical expression is as follows:
[0028]
[0029] Among them, C1~C3 represent the equivalent heat capacity of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, respectively, T h Indicates the hot end temperature of the thermoelectric generator, T c Indicates the cold end temperature of the thermoelectric generator, T sIndicates the heat sink temperature, T a represents the ambient temperature, θ c1 Represents the equivalent thermal resistance between the cold and hot ends of the thermoelectric generator, θ c2 Represents the equivalent thermal resistance between the cold end of the thermoelectric generator and the heat sink, θ v represents the equivalent convection thermal resistance between the heat sink and the environment, Q in It represents the total heat absorbed by the hot end of the thermoelectric generator. Q1 represents the heat absorbed from the hot end, Q2 represents the heat released from the cold end to the radiator, and Q j P is the irreversible Joule heat generated inside the thermoelectric generator, half of which flows to the hot end and the other half flows to the cold end. e It is the power output by the thermoelectric generator.
[0030] In the formula, C1, C2, C3, θ c1 ,θ c2 ,θ v , Q1, Q2, Q j , P e Determined by the following formulas:
[0031] C i =c i m i
[0032]
[0033]
[0034]
[0035]
[0036]
[0037]
[0038] In the formula, the subscript i∈{1,2,3} represents the hot end, cold end or heat sink of the thermoelectric generator, c i and m i are the specific heat capacity and mass of the material, d i is the thermal conduction distance, A i is the heat conduction area, λ i is the thermal conductivity, A s is the convective heat transfer area of the radiator, h s is the convective heat transfer coefficient, α pn is the Seebeck coefficient between P-type and N-type semiconductor thermoelectric elements, d is the total number of pairs of semiconductor thermoelectric elements, I o is the current formed in the loop, R s is the internal resistance of the device, RL Connect the load to the thermoelectric generator.
[0039] According to the above formula, it can be found that the equivalent thermal network is essentially a nonlinear system, so the temperature of the thermoelectric generator can be regarded as the dynamic response of the system. However, since the traditional equivalent thermal network ignores the nonlinear heat exchange between the thermoelectric generator and the surrounding environment, it is difficult to accurately obtain the dynamic change of the temperature difference between the cold and hot ends. In order to enable the traditional equivalent thermal network to describe the nonlinear heat dissipation characteristics of the thermoelectric generator, it needs to be improved.
[0040] The improved equivalent thermal network model in step 1 is as follows: Figure 3 As shown in the figure, the basic idea of improvement is to introduce nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator into the external thermal network respectively. Through time domain discretization, the mathematical expression of the improved equivalent thermal network model is as follows:
[0041] Through time domain discretization, the improved mathematical expression of the equivalent thermal network model is as follows:
[0042]
[0043] Where k represents the time, Δt represents the discrete step length; T h Indicates the hot end temperature of the thermoelectric generator, T c Indicates the cold end temperature of the thermoelectric generator, T s Indicates the heat sink temperature, T a Indicates ambient temperature; Q i Represents the heat value, i = 1, 2, Q1 represents the heat absorbed from the hot end, Q2 represents the heat released from the cold end to the radiator; α pn Represents the Seebeck coefficient between P-type and N-type semiconductor thermoelectric elements, θ c1 Represents the equivalent thermal resistance between the cold and hot ends of the thermoelectric generator, θ c2 It represents the equivalent thermal resistance between the cold end of the thermoelectric generator and the heat sink; I o Indicates the current formed in the loop, I L represents the current in the thermoelectric generator, R s represents the internal resistance of the device, a1~a3 represent the proportional coefficients of the equivalent convection thermal resistance of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, b1~b3 represent the exponential coefficients of the equivalent convection thermal resistance of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, δ represents the Stepan-Boltzmann constant, ε1~ε3 represent the emissivity of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, A r1 ~A r2 represents the radiation heat transfer area of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, C1~C3 represent the equivalent heat capacity of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, Pe Indicates the power generated by the thermoelectric generator.
[0044] Step 2: Based on the improved equivalent thermal network model, a multi-objective function is constructed, and the thermal parameters of the thermoelectric generator are extracted by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results.
[0045] Step 2-1) Taking the minimum error between the calculated temperature results and the experimental results of the three temperatures of the hot end, cold end and heat sink of the thermoelectric generator as the identification target, a multi-objective function for parameter identification is constructed, as follows:
[0046] Assume that the calculated values of the hot end, cold end and heat sink temperature of the thermoelectric generator at time k of the nth group of experiments are and Their corresponding measured values are and
[0047] Where J1(X)~J3(X) represent the multi-objective function expressions of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, respectively, and X is the quantity to be identified.
[0048] Under the specified heat source power and ambient temperature, a thermoelectric power generation experiment was carried out. The appropriate ambient temperature and experimental duration were set. The thermal parameters of the thermoelectric generator were extracted by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results in the improved equivalent thermal network.
[0049] Step 2-2) Combine the experimental sample data and the NSGA-II algorithm to obtain the thermal parameters of the thermoelectric generator and obtain the parameter identification results, as follows:
[0050] exist Figure 4 Five thermoelectric power generation experiments were carried out under the heat source power shown. During the experiment, the ambient temperature was maintained at about 20°C and the experiment lasted for 100 minutes. The thermal parameters of the thermoelectric generator were extracted by combining the experimental sample data and the NSGA-II algorithm, and the parameter identification results in the improved equivalent thermal network were obtained, as shown in Figure 2. Figure 5 shown.
[0051] Step 3: Using the improved equivalent thermal network model and identification results, under given working conditions, estimate the temperature of the thermoelectric generator to obtain the estimated temperature difference between the cold and hot ends.
[0052] Combination Figure 6 The heat source power and ambient temperature are used to estimate the cold and hot end temperatures of the thermoelectric generator. At the same time, the traditional equivalent thermal network is used to estimate the temperature of the thermoelectric generator, and the estimated values and measured values of the hot end, cold end and heat sink temperatures obtained by the two methods are obtained, as shown in the figure. Figure 7 , 8 , as shown in Figure 9.
[0053] The estimated error of the traditional equivalent thermal network model is calculated based on the obtained data. Fig.10 As shown in the figure, the estimation error of the improved equivalent thermal network model is as follows: Fig.11 As shown in the figure. The maximum estimation error of the traditional equivalent thermal network model can be as high as 20%, and the average error exceeds 20.09%, while the maximum estimation error of the improved equivalent thermal network model is only 9.18%, 4.92% and 8.78%, and the average error is only 1.87%, 2.04 and 2.8%. This shows that the estimation results of the proposed method are more consistent with the measured results.
[0054] The above results verify the effectiveness and precision advantages of the proposed method, and provide an important basis for improving the accuracy of performance evaluation of thermoelectric power generation systems and the reliability of power generation prediction.
Claims
1. A method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model, characterized in that: The traditional equivalent thermal network model is improved, and nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator are introduced into the external thermal network of the traditional equivalent thermal network model respectively; and the thermal parameters in the thermal network are extracted by multi-objective identification and NSGA-II algorithm, so as to realize the rapid estimation of the temperature difference between the hot and cold ends of the thermoelectric generator.
2. The method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model according to claim 1, characterized in that: The following steps are involved: Step 1: According to the nonlinear principle of radiation heat dissipation and convection heat dissipation, based on the traditional equivalent thermal network model, nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator are introduced into the external thermal network of the traditional equivalent thermal network model to design an improved equivalent thermal network model; Step 2: construct a multi-objective function based on the improved equivalent thermal network model, and extract the thermal parameters of the thermoelectric generator by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results; Step 3: Using the improved equivalent thermal network model and identification results, under given working conditions, estimate the temperature of the thermoelectric generator to obtain the estimated temperature difference between the cold and hot ends.
3. The method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model according to claim 2, characterized in that: In step 1, according to the nonlinear principle of radiation heat dissipation and convection heat dissipation, based on the traditional equivalent thermal network model, nonlinear heat transfer branches of the hot end, cold end and radiator of the thermoelectric generator are introduced into the external thermal network of the traditional equivalent thermal network model, and the improved equivalent thermal network model is designed as follows: Through time domain discretization, the improved mathematical expression of the equivalent thermal network model is as follows: Where k represents the time, Δt represents the discrete step length; T h Indicates the hot end temperature of the thermoelectric generator, T c Indicates the cold end temperature of the thermoelectric generator, T s Indicates the heat sink temperature, T a Indicates ambient temperature; Q i Represents the heat value, i = 1, 2, Q1 represents the heat absorbed from the hot end, Q2 represents the heat released from the cold end to the radiator; α pn Represents the Seebeck coefficient between P-type and N-type semiconductor thermoelectric elements, θ c1 Represents the equivalent thermal resistance between the cold and hot ends of the thermoelectric generator, θ c2 It represents the equivalent thermal resistance between the cold end of the thermoelectric generator and the heat sink; I o Indicates the current formed in the loop, I L represents the current in the thermoelectric generator, R s represents the internal resistance of the device, a1~a3 represent the proportional coefficients of the equivalent convection thermal resistance of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, b1~b3 represent the exponential coefficients of the equivalent convection thermal resistance of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, δ represents the Stepan-Boltzmann constant, ε1~ε3 represent the emissivity of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the heat sink, A r1 ~A r2 represents the radiation heat transfer area of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, C1~C3 represent the equivalent heat capacity of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator, P e Indicates the power generated by the thermoelectric generator.
4. The method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model according to claim 2, characterized in that: In step 2, a multi-objective function is constructed based on the improved equivalent thermal network model, and the thermal parameters of the thermoelectric generator are improved by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results, as follows: Taking the minimum error between the calculated results and the experimental results of the three temperatures of the hot end, cold end and radiator of the thermoelectric generator as the identification target, a multi-objective function for parameter identification is constructed. The mathematical expression is as follows: Assume that the calculated values of the hot end, cold end and heat sink temperature of the thermoelectric generator at time k of the nth group of experiments are and Their corresponding measured values are and In the formula, J1(X)~J3(X) represent the multi-objective function expressions of the hot end of the thermoelectric generator, the cold end of the thermoelectric generator and the radiator respectively, and X is the quantity to be identified; Under the specified heat source power and ambient temperature, a thermoelectric power generation experiment was carried out. The appropriate ambient temperature and experimental duration were set. The thermal parameters of the thermoelectric generator were extracted by combining the experimental sample data and the NSGA-II algorithm to obtain the parameter identification results in the improved equivalent thermal network.
5. The method for estimating the temperature of a thermoelectric power generation module based on an improved thermal network model according to claim 4, characterized in that: In step 3, the temperature of the thermoelectric generator is estimated by using the improved equivalent thermal network model and parameter identification results under the specified heat source power and environmental conditions, so as to obtain the estimated result of the temperature difference between the cold and hot ends, as follows: Under the specified heat source power and ambient temperature, the hot and cold end temperatures of the thermoelectric generator are estimated. For the sake of comparison, the traditional equivalent thermal network is also used to estimate the hot and cold end temperatures of the thermoelectric generator, and the estimated and measured values of the hot end, cold end and heat sink temperatures obtained by the two methods are obtained. The estimation result errors of the traditional equivalent thermal network model and the improved equivalent thermal network model were calculated based on the obtained data. The maximum estimation error and the average error of the data obtained by the traditional equivalent thermal network model exceeded those of the improved equivalent thermal network model.
