A long-cycle modular high-heat-density solid thermal storage battery and its energy conversion model

Abstract

The present invention discloses a long-period modular high-heat-density solid thermal storage battery and its energy conversion model, which belong to the field of power systems. The electric heat exchange device provides thermal power for the high-heat-density medium to store heat, and the high-heat-density medium is used to store the thermal energy generated by the electric heat exchange device; the thermoelectric conversion device converts the thermal energy stored in the high-heat-density medium into electrical energy, and the thermal insulation layer is used to isolate the heat transfer between the high-heat-density medium and the outside air. Based on the LSTM algorithm, the present invention proposes an LSTM-based electric-thermal conversion efficiency model and an electric-thermal conversion efficiency model, which are respectively used to improve the accuracy of the calculation of electric-thermal conversion efficiency and thermal-electric conversion efficiency. The present invention has the advantages of low manufacturing difficulty and low cost, and can be modularly assembled according to the needs of different scenarios (new energy capacity differences, regional load differences, heating demand differences, etc.), and has broad application prospects.

Description

Technical Field

[0001] The present invention belongs to the field of electric power systems, and in particular relates to a long-period modular high-heat-density solid thermal storage battery and an energy conversion model thereof. Background Art

[0002] The long-term intermittent, volatility and uncertainty of wind power and photovoltaics pose severe challenges to the operation of new power systems. Energy storage can store wind and photovoltaic power when it is difficult to absorb it, and discharge to supplement electricity when there is a shortage of wind and photovoltaic power, thereby alleviating the impact of wind power and photovoltaic power fluctuations, long-term intermittent, etc. on the operation of the power grid. Actively promoting the research and construction of energy storage is one of the effective measures for the power grid to cope with the above-mentioned operational challenges. At present, energy storage includes pumped storage and new energy storage. Pumped storage has high construction costs and long construction periods, and it is difficult to effectively cope with the operational challenges faced by the power grid in the short term. New energy storage mainly includes electrochemical energy storage, compressed air energy storage, flywheel energy storage, etc., but the above-mentioned new energy storage methods currently have high construction costs and are difficult to promote on a large scale. The battery capacity is limited and it is difficult to cope with the long-term intermittent nature of wind power and photovoltaics. In addition, different scenarios have different requirements for battery capacity. If the battery capacity is too large, the investment cost will be too high and the battery utilization rate will be low. If the battery capacity is too small, it will be difficult to meet the needs of grid operation. Therefore, it is crucial to adapt to local conditions and build suitable energy storage capacity, modular construction and installation of battery energy storage according to the needs of different scenarios. Summary of the invention

[0003] In view of the above problems, the present invention provides a long-cycle modular high-heat-density solid thermal storage battery and its energy conversion model, which are used to solve the problems of excessively high investment costs and low battery utilization due to the existing energy storage methods.

[0004] The technical solution adopted by the present invention is as follows: a long-period modular high-heat-density solid heat storage battery, comprising an electric heat exchange device, a high-heat-density medium, a thermoelectric conversion device and a thermal insulation layer, wherein the electric heat exchange device provides thermal power for the high-heat-density medium to store heat, and the high-heat-density medium is used to store the thermal energy generated by the electric heat exchange device; the thermoelectric conversion device converts the thermal energy stored in the high-heat-density medium into electrical energy, and the thermal insulation layer is used to isolate the heat transfer between the high-heat-density medium and the outside air.

[0005] Furthermore, the heat density calculation method of the high heat density solid thermal storage battery is shown in formula (1):

[0006] ρ r =ρ m ×C (1)

[0007] In the formula, ρ r is the thermal density of the material, and its physical meaning is per unit volume 1m 3The amount of heat that a material can store under a unit temperature change of 1°C, expressed in kJ / (m 3 ·℃); ρ m is the density of the material; C is the specific heat capacity of the material.

[0008] Furthermore, the maximum heat storage capacity Q of the high heat density solid thermal storage battery max The calculation method of is shown in formula (2).

[0009] Q max =ρ r ×V×ΔT (2)

[0010] Where V is the volume of the heat-tight material; ΔT is the maximum temperature difference that the heat-tight material can be heated to.

[0011] Furthermore, the heating methods of the electric heat exchange device include resistance heating, electromagnetic induction heating and microwave heating.

[0012] Furthermore, the thermoelectric conversion device includes a thermoelectric generator, a heat pump and a thermal steam turbine generator.

[0013] Furthermore, the high heat density medium includes steel, iron, concrete, asbestos and ceramics.

[0014] Furthermore, the thermal insulation layer includes glass fiber, rock wool, aluminum silicate fiber, polyurethane foam, foam glass and diatomaceous earth.

[0015] Another object of the present invention is to provide a method for constructing an electric-thermal conversion model based on the above-mentioned long-period modular high-heat-density solid thermal storage battery, as follows: the high-heat-density solid thermal storage battery converts the input electric power P 1 Converted into heat power h 1 The total thermal power of the thermoelectric conversion device is h z , where the thermal power used for heating is h 3 , the thermal power converted into electrical energy is h 2 , the output power is P 2 ; Formula (3) is the LSTM-based electric-thermal conversion efficiency model. The LSTM algorithm is used to train the injection power and thermal power of historical high-density solid thermal storage batteries under different working conditions to obtain more accurate dynamic electric-thermal conversion efficiency under different working conditions. Formula (4) is the heat calculation method of high-density solid thermal storage batteries. The residual heat Q(t 0 ) and the newly injected thermal power generate heat. Formula (5) is the electric-thermal conversion constraint condition of high thermal density solid thermal storage battery, including thermal storage capacity constraint, electric power constraint and thermal power constraint, to ensure that the thermal storage capacity, electric power and thermal power are all operated within the constraint range to ensure the safety of the thermal storage battery;

[0016] h 1 (t) = λ(t)·P 1 (t) (3)

[0017]

[0018]

[0019] In the formula, h 1 (t) is the converted thermal power of the high heat density solid thermal storage battery electric-heat conversion model at time t; P 1 (t) is the converted electric power of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; λ(t) is the electric-thermal conversion efficiency of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t) is the heat storage capacity of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t 0 ) is t 0 The heat storage of the electric-thermal conversion model of high thermal density solid thermal storage battery at the moment; Q max and Q min are the maximum and minimum heat storage capacity of the electric-thermal conversion model of high thermal density solid thermal storage battery; P 1,max and P 1,min are the maximum and minimum electric powers of the electric-thermal conversion model of high thermal density solid thermal storage battery; h 1,max and h 1,min They are the maximum and minimum thermal powers of the electric-thermal conversion model of high thermal density solid thermal storage battery;

[0020] The thermal power, electric power, and electric-to-thermal conversion efficiency data under different historical working conditions are used as input samples, where the input sample is: X(t) = [λ(t), h 1 (t),P 1 (t)], the output is: F C (t+Δt)=[λ(t+Δt)], the standard electrothermal conversion efficiency is input into the hidden layer, and then the LSTM deep learning network model is trained based on the input sample data and cell state information. The training process is shown in formula (6).

[0021]

[0022] Where, σ is the activation function of the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; f t is the output of the forget gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W f 、b f is the forget gate matrix of the corresponding high thermal density solid thermal storage battery electric-thermal conversion efficiency model; i t The output of the input gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W i、b i is the input gate weight matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; C t-1 Model old cell state information for high heat density solid thermal storage battery electric-thermal conversion efficiency; Candidate state information added for the selection of the electric-thermal conversion efficiency model of high thermal density solid thermal storage batteries; C t Updated cell information for the electric-thermal conversion efficiency model of high heat density solid thermal storage batteries; W C 、b C The neuron matrix corresponding to the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; t The output of the output gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W o 、b o The output gate matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; e t is the output result of the electric-thermal conversion efficiency model of high heat density solid thermal storage battery; the electric-thermal conversion efficiency model of high heat density solid thermal storage battery is shown in formula (7):

[0023] λ(t+Δt)=f λ (λ(t),h 1 (t),P 1 (t)) (7)

[0024] In the formula, f λ () is the LSTM prediction model for the electric-thermal conversion efficiency of high thermal density thermal storage batteries. According to this prediction model, when the input λ(t), h 1 (t),P 1 (t), the battery electric-thermal conversion efficiency λ(t+Δt) at the future time t+Δt can be predicted.

[0025] Another object of the present invention is to provide a method for constructing a heat-to-electric conversion model of a long-period modular high-heat-density solid thermal storage battery based on the above-mentioned method, as follows: Formulas (8) and (9) are heat-to-electric conversion efficiency models of high-heat-density solid thermal storage batteries based on LSTM, Formula (10) is a thermal balance power model in the heat-to-electric conversion process of high-heat-density solid thermal storage batteries, Formula (11) is a residual heat calculation model in the heat-to-electric conversion process of high-heat-density solid thermal storage batteries, Formula (12) is a constraint condition in the heat-to-electric conversion process of high-heat-density solid thermal storage batteries,

[0026] P 2 (t) = ξ(t)·h 2 (t) (8)

[0027] ξ(t)=f ξ (ξ(t-Δt),h 2 (t-Δt),P 2(t-Δt)) (9)

[0028] h z (t) = h 2 (t)+h 3 (t) (10)

[0029]

[0030]

[0031] Where P 2 (t) is the output power of the high heat density solid thermal storage battery at time t; ξ(t) is the heat-to-electricity conversion efficiency of the high heat density solid thermal storage battery trained based on LSTM; h 2 (t) is the thermal power converted into electrical energy by the high heat density solid thermal storage battery at time t; f ξ () is the LSTM prediction model for the thermoelectric conversion efficiency of high heat storage battery. According to this model, when the input ξ(t-Δt),h 2 (t-Δt),P 2 (t-Δt), the battery electric-thermal conversion efficiency ξ(t) at the next time t can be predicted; h z (t) is the total conversion heat power during the thermoelectric conversion process of the high heat density solid thermal storage battery at time t; h 3 (t) is the heat power provided by the high heat density solid thermal storage battery during the thermoelectric conversion process at time t; Q S (t) is the residual heat capacity of the high heat density solid thermal storage battery at time t; Q(t 1 ) is t 1 Thermal capacity of high thermal density solid thermal storage battery at all times; Q min P is the minimum heat storage capacity of the thermal storage battery; 2,max and P 2,min are the maximum and minimum electric power of the thermal-electric conversion model of the thermal storage battery; h 2,max and h 2,min are the maximum and minimum thermal powers converted into electrical energy by the thermal-electric conversion model of the thermal storage battery; h 3,max and h 3,min are the maximum and minimum heating powers of the thermal-electric conversion model of the thermal storage battery; h z,max and h z,min are the total maximum and minimum thermal powers of the thermal-electric conversion model of the thermal storage battery, respectively.

[0032] Advantages and beneficial effects of the present invention: The present invention converts electrical energy into thermal energy through a thermoelectric conversion device, and stores the thermal energy in a high heat density medium. Based on the characteristics of high heat density medium, which has high heat storage density, can be heated in a large temperature range (maximum 0 to 1500°C), and has a large heat storage potential, a large amount of electrical energy such as new energy can be stored. The electric-thermal conversion efficiency model and the electric-thermal conversion efficiency model constructed by the present invention are used to improve the accuracy of the calculation of electric-thermal conversion efficiency and thermal-electric conversion efficiency, respectively. The present invention has the advantages of low manufacturing difficulty and low cost, and can be modularly assembled according to the needs of different scenarios (new energy capacity differences, regional load differences, heating demand differences, etc.). The stored electrical energy includes wind power and photovoltaic energy, which can promote the consumption of new energy and balance the power of the power grid, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Structural schematic diagram of long-cycle modular high-heat-density solid thermal storage battery;

[0034] Figure 2 This is a model diagram of the electric-thermal conversion efficiency of high thermal density solid thermal storage batteries based on LSTM. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is described below by the specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0036] Example 1

[0037] like Figure 1 As shown, a long-period modular high heat density solid thermal storage battery includes an electric heat exchange device 1, a high heat density medium 2, a thermoelectric conversion device 3 and a heat insulation layer 4. The electric heat exchange device 1 converts the input electric power P 1 Converted into heat power h 1 , providing thermal power for the high heat density medium 2 to store heat; the high heat density medium 2 is used to store the heat energy generated by the electric heat exchange device 1, and the maximum stored heat is Q max The thermoelectric conversion device 3 converts the thermal energy stored in the high heat density medium 2 into electrical energy. The total thermal power of the thermoelectric conversion device 3 is h z , where the thermal power used for heating is h 3 , the thermal power converted into electrical energy is h 2 , the output power is P 2; The thermal insulation layer 4 is used to isolate the heat transfer between the high heat density medium 2 and the outside air, prevent the heat stored in the high heat density medium 2 from being dissipated into the outside air, and thus realize the long-term storage of heat. The heating methods of the electric heat exchange device 1 include resistance heating, electromagnetic induction heating and microwave heating. The thermoelectric conversion device 3 includes a thermoelectric generator, a heat pump and a steam turbine generator. The high heat density medium 2 includes steel, iron, concrete, asbestos and ceramics, which can be selected according to the actual materials available on site, or mixed. The medium must have the characteristics of low material cost, convenient material acquisition, easy implementation, large heat storage potential, good thermal conductivity, etc., and can realize the long-term storage of different forms of energy and electricity such as new energy. The thermal insulation layer 4 includes glass fiber, rock wool, aluminum silicate fiber, polyurethane foam, foam glass and diatomaceous earth.

[0038] This embodiment converts electrical energy into thermal energy through a thermoelectric conversion device, and stores the thermal energy in a high heat density medium. Based on the characteristics of high heat density medium, which has high heat storage density, can be heated in a large temperature range (maximum 0 to 1500°C), and has a large heat storage potential, a large amount of electrical energy such as new energy can be stored. Combined with the thermal insulation layer, long-term heat storage can be achieved. The thermoelectric conversion device releases the heat stored in the high heat density medium, converts it into electrical energy, and can output thermal energy to provide heating to users.

[0039] The heat density calculation method of the high heat density solid thermal storage battery is shown in formula (1):

[0040] ρ r =ρ m ×C (1)

[0041] In the formula, ρ r is the thermal density of the material, and its physical meaning is per unit volume 1m 3 The amount of heat that a material can store under a unit temperature change of 1°C, expressed in kJ / (m 3 ·℃); ρ m is the density of the material; C is the specific heat capacity of the material.

[0042] The maximum heat storage capacity calculation method of the high heat density solid thermal storage battery is shown in formula (2):

[0043] Q max =ρ r ×V×ΔT (2)

[0044] In the formula, ρ r is the thermal density of the material; V is the volume of the heat-tight material; ΔT is the maximum temperature difference that the heat-tight material can be heated to.

[0045] Example 2

[0046] This embodiment provides an electric-thermal conversion model of a long-period modular high-heat-density solid thermal storage battery based on embodiment 1, including an electric-thermal conversion efficiency model of a high-heat-density solid thermal storage battery based on LSTM, a heat calculation method for a high-heat-density solid thermal storage battery, and electric-thermal conversion constraints for a high-heat-density solid thermal storage battery. Among them, the electric-thermal conversion efficiency model aims at the difference of electric-thermal conversion efficiency under different operating modes or powers. In order to improve the calculation accuracy, the LSTM algorithm is used to train the injected power and thermal power under different working conditions of the historical thermal storage battery to obtain a more accurate dynamic electric-thermal conversion efficiency under different working conditions. The heat calculation method mainly calculates the current thermal capacity of the battery based on the heat generated by the residual heat and the newly injected thermal power. The electric-thermal conversion constraints include thermal storage capacity constraints, electric power constraints, and thermal power constraints to ensure that the thermal storage capacity, electric power, and thermal power are all operated within the constraints, and to ensure the safety of the thermal storage battery during the electric-thermal conversion process.

[0047] The construction method is as follows: As shown in formulas (3)-(5), formula (3) is the LSTM-based electric-thermal conversion efficiency model, and the training process is as follows: Figure 2 As shown in the figure, the LSTM algorithm is used to train the injection power and thermal power of the historical high heat density solid thermal storage battery under different working conditions to obtain a more accurate dynamic electric-thermal conversion efficiency under different working conditions. Formula (4) is the heat calculation method of the high heat density solid thermal storage battery. The residual heat Q(t 0 ) and the newly injected thermal power generate heat. Formula (5) is the electric-thermal conversion constraint condition of high thermal density solid thermal storage battery, including thermal storage capacity constraint, electric power constraint and thermal power constraint, to ensure that the thermal storage capacity, electric power and thermal power are all operated within the constraint range to ensure the safety of the thermal storage battery;

[0048] h 1 (t) = λ(t)·P 1 (t) (3)

[0049]

[0050]

[0051] In the formula, h 1 (t) is the converted thermal power of the high heat density solid thermal storage battery electric-heat conversion model at time t; P 1 (t) is the converted electric power of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; λ(t) is the electric-thermal conversion efficiency of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t) is the heat storage capacity of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t 0 ) is t 0 The heat storage of the electric-thermal conversion model of high thermal density solid thermal storage battery at the moment; Q max and Qmin are the maximum and minimum heat storage capacity of the electric-thermal conversion model of high thermal density solid thermal storage battery; P 1,max and P 1,min are the maximum and minimum electric powers of the electric-thermal conversion model of high thermal density solid thermal storage battery; h 1,max and h 1,min They are the maximum and minimum thermal powers of the electric-thermal conversion model of high thermal density solid thermal storage battery;

[0052] The thermal power, electric power, and electric-to-thermal conversion efficiency data under different historical working conditions are used as input samples, where the input sample is: X(t) = [λ(t), h 1 (t),P 1 (t)], the output is: F C (t+Δt)=[λ(t+Δt)], the standard electrothermal conversion efficiency is input into the hidden layer, and then the LSTM deep learning network model is trained based on the input sample data and cell state information. The training process is shown in formula (6).

[0053]

[0054] Where, σ is the activation function of the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; f t is the output of the forget gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model, W f 、b f is the forget gate matrix of the corresponding high thermal density solid thermal storage battery electric-thermal conversion efficiency model; i t is the output of the input gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model, W i 、b i is the input gate weight matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; C t-1 Model old cell state information for high heat density solid thermal storage battery electric-thermal conversion efficiency, Candidate state information added for the model selection of high heat density solid thermal storage battery electric-thermal conversion efficiency, C t Updated cell information for the electric-thermal conversion efficiency model of high heat density solid thermal storage batteries, W C 、b C The neuron matrix corresponding to the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; t is the output of the output gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model, W o 、b o The output gate matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; e tis the output result of the high heat density solid thermal storage battery electric-thermal conversion efficiency model. The final high heat density solid thermal storage battery electric-thermal conversion efficiency model is shown in formula (7):

[0055] λ(t+Δt)=f λ (λ(t),h 1 (t),P 1 (t)) (7)

[0056] In the formula, f λ () is the LSTM prediction model for the electric-thermal conversion efficiency of high thermal density thermal storage batteries. According to this prediction model, when the input λ(t), h 1 (t),P 1 (t), the battery electric-thermal conversion efficiency λ(t+Δt) at the future time t+Δt can be predicted.

[0057] Example 3

[0058] This embodiment provides a heat-to-electric conversion model of a long-period modular high-heat-density solid thermal storage battery based on embodiment 1, including a heat-to-electric conversion efficiency model of a high-heat-density solid thermal storage battery based on LSTM, a heat balance power model in the heat-to-electric conversion process, a residual heat calculation model in the heat-to-electric conversion process, and heat-to-electric conversion constraints. Among them, the heat-to-electric conversion efficiency model is aimed at the differences in the heat-to-electric conversion efficiency under different operating modes or powers. In order to improve the calculation accuracy, the output power and thermal power of the historical thermal storage battery under different working conditions are trained using the LSTM algorithm to obtain a more accurate dynamic thermoelectric conversion efficiency under different working conditions. The thermal balance power model mainly guarantees the balance of thermal power during the thermoelectric conversion process. The residual heat calculation model in the heat-to-electric conversion process mainly calculates the residual heat of the battery after the battery is converted into electrical energy or heat supply. The heat-to-electric conversion constraints mainly ensure that the thermal power, electrical power, and battery thermal capacity of the battery are within the safe operating range during the thermoelectric conversion process, and ensure the safety of the heat-to-electric conversion of the thermal storage battery.

[0059] The construction method is as follows: Equations (8) and (9) are the heat-to-electric conversion efficiency models of high heat density solid thermal storage batteries based on LSTM. The LSTM algorithm is used to train the output power and thermal power of historical high heat density solid thermal storage batteries under different working conditions to obtain more accurate dynamic heat-to-electric conversion efficiency under different working conditions. Equation (10) is the thermal balance power model in the heat-to-electric conversion process. Equation (11) is the residual heat calculation model in the heat-to-electric conversion process. Equation (12) is the constraint condition in the heat-to-electric conversion process.

[0060] P 2 (t) = ξ(t)·h 2 (t) (8)

[0061] ξ(t)=f ξ(ξ(t-Δt),h 2 (t-Δt),P 2 (t-Δt)) (9)

[0062] h z (t) = h 2 (t)+h 3 (t) (10)

[0063]

[0064]

[0065] Where P 2 (t) is the output power of the high heat density solid thermal storage battery at time t; ξ(t) is the heat-to-electricity conversion efficiency of the high heat density solid thermal storage battery trained based on LSTM; h 2 (t) is the thermal power converted into electrical energy by the high heat density solid thermal storage battery at time t; f ξ () is the LSTM prediction model for the thermoelectric conversion efficiency of high heat storage battery. According to this model, when the input ξ(t-Δt),h 2 (t-Δt),P 2 (t-Δt), the battery electric-thermal conversion efficiency ξ(t) at the next time t can be predicted; h z (t) is the total conversion heat power during the thermoelectric conversion process of the high heat density solid thermal storage battery at time t; h 3 (t) is the heat power provided by the high heat density solid thermal storage battery during the thermoelectric conversion process at time t; Q S (t) is the residual heat capacity of the high heat density solid thermal storage battery at time t; Q(t 1 ) is t 1 Thermal capacity of high thermal density solid thermal storage battery at all times; Q min P is the minimum heat storage capacity of the thermal storage battery; 2,max and P 2,min are the maximum and minimum electric power of the thermal-electric conversion model of the thermal storage battery; h 2,max and h 2,min are the maximum and minimum thermal powers converted into electrical energy by the thermal-electric conversion model of the thermal storage battery; h 3,max and h 3,min are the maximum and minimum heating powers of the thermal-electric conversion model of the thermal storage battery; h z,max and h z,min are the total maximum and minimum thermal powers of the thermal-electric conversion model of the thermal storage battery, respectively. Therefore, based on the above process, the conversion of thermal energy to electrical energy of a high thermal density thermal storage battery can be realized.

Claims

1. An electric-heat conversion model of a long-period modular high-heat-density solid thermal storage battery, characterized in that: Constructed by the following method: A long-cycle modular high heat density solid heat storage battery is adopted, which includes an electric heat exchange device, a high heat density medium, a thermoelectric conversion device and a heat insulation layer, wherein the electric heat exchange device provides thermal power for the high heat density medium to store heat, and the high heat density medium is used to store the heat energy generated by the electric heat exchange device; the thermoelectric conversion device converts the heat energy stored in the high heat density medium into electrical energy, and the heat insulation layer is used to isolate the heat transfer between the high heat density medium and the outside air; wherein the heating method of the electric heat exchange device includes resistance heating, electromagnetic induction heating and microwave heating; the thermoelectric conversion device includes a thermoelectric generator, a heat pump and a steam turbine generator; the high heat density medium includes steel, iron, concrete, asbestos and ceramics; the heat insulation layer includes glass fiber, rock wool, aluminum silicate fiber, polyurethane foam, foam glass and diatomaceous earth; wherein the heat density calculation method of the high heat density solid heat storage battery is as shown in formula (1), r r =ρ m ×C (1) In the formula, ρ r is the thermal density of the material, and its physical meaning is per unit volume 1m 3 The amount of heat that a material can store under a unit temperature change of 1°C, expressed in kJ / (m 3 ·℃); ρ m is the density of the material; C is the specific heat capacity of the material; Among them, the maximum heat storage capacity Q of the high heat density solid thermal storage battery max The calculation method is shown in formula (2): Q max =ρ r ×V×ΔT (2) Where V is the volume of the heat-tight material; ΔT is the maximum temperature difference that the heat-tight material can be heated; The high heat density solid thermal storage battery converts the input electric power P1 into thermal power h1; the total thermal power of the thermoelectric conversion device is h1. z , where the thermal power used for heating is h3, the thermal power converted to electrical energy is h2, and the output electrical energy is P2; Formula (3) is an LSTM-based electric-thermal conversion efficiency model. The LSTM algorithm is used to train the injected power and thermal power of historical high-density solid thermal storage batteries under different working conditions to obtain more accurate dynamic electric-thermal conversion efficiency under different working conditions. Formula (4) is a heat calculation method for high-density solid thermal storage batteries, which is composed of residual heat Q(t0) and heat generated by newly injected thermal power. Formula (5) is the electric-thermal conversion constraint condition of high-density solid thermal storage batteries, including heat storage capacity constraint, electric power constraint and thermal power constraint, to ensure that the heat storage capacity, electric power and thermal power are all within the constraint range to ensure the safety of the thermal storage battery; h1(t)=λ(t)·P1(t) (3) Wherein, h1(t) is the converted thermal power of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; P1(t) is the converted electric power of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; λ(t) is the electric-thermal conversion efficiency of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t) is the heat storage capacity of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t; Q(t0) is the heat storage capacity of the electric-thermal conversion model of the high heat density solid thermal storage battery at time t0; Q max and Q min are the maximum and minimum heat storage capacity of the electric-thermal conversion model of high thermal density solid thermal storage battery; P 1,max and P 1,min are the maximum and minimum electric powers of the electric-thermal conversion model of high thermal density solid thermal storage battery; h 1,max and h 1,min are the maximum and minimum thermal powers of the electric-thermal conversion model of high thermal density solid thermal storage battery; The thermal power, electric power, and electric-to-heat conversion efficiency data under different historical working conditions are used as input samples, where the input sample is: X(t) = [λ(t), h1(t), P1(t)], and the output is: F C (t+Δt)=[λ(t+Δt)], the standard electrothermal conversion efficiency is input into the hidden layer, and then the LSTM deep learning network model is trained based on the input sample data and cell state information. The training process is shown in formula (6). Where, σ is the activation function of the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; f t is the output of the forget gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W f 、b f is the forget gate matrix of the corresponding high thermal density solid thermal storage battery electric-thermal conversion efficiency model; i t The output of the input gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W i 、b i is the input gate weight matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; C t-1 Model old cell state information for high heat density solid thermal storage battery electric-thermal conversion efficiency; Candidate state information added for the selection of the electric-thermal conversion efficiency model of high thermal density solid thermal storage batteries; C t Updated cell information for the electric-thermal conversion efficiency model of high heat density solid thermal storage batteries; W C 、b C The neuron matrix corresponding to the electric-thermal conversion efficiency model of high thermal density solid thermal storage battery; t The output of the output gate of the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; W o 、b o The output gate matrix corresponding to the high thermal density solid thermal storage battery electric-thermal conversion efficiency model; e t is the output result of the electric-thermal conversion efficiency model of high heat density solid thermal storage battery; the electric-thermal conversion efficiency model of high heat density solid thermal storage battery is shown in formula (7): λ(t+Δt)=f λ (λ(t),h1(t),P1(t)) (7) In the formula, f λ () is the LSTM prediction model for the electrothermal conversion efficiency of high heat storage batteries. According to this prediction model, when λ(t), h1(t), and P1(t) are input, the battery electrothermal conversion efficiency λ(t+Δt) at the future time t+Δt can be predicted.

2. The heat-to-electricity conversion model of a long-period modular high-heat-density solid thermal storage battery according to claim 1, characterized in that: Formulas (8) and (9) are the heat-to-electric conversion efficiency models of high heat density solid thermal storage batteries based on LSTM. Formula (10) is the thermal balance power model in the heat-to-electric conversion process of high heat density solid thermal storage batteries. Formula (11) is the residual heat calculation model in the heat-to-electric conversion process of high heat density solid thermal storage batteries. Formula (12) is the constraint condition in the heat-to-electric conversion process of high heat density solid thermal storage batteries. P2(t)=ξ(t)·h2(t) (8) ξ(t)=f ξ (ξ(t-Δt),h2(t-Δt),P2(t-Δt)) (9) h z (t)=h2(t)+h3(t) (10) Where P2(t) is the output power of the high heat density solid thermal storage battery at time t; ξ(t) is the heat-to-electric conversion efficiency of the high heat density solid thermal storage battery trained based on LSTM; h2(t) is the thermal power converted into electrical energy by the high heat density solid thermal storage battery at time t; f ξ () is the LSTM prediction model for the thermoelectric conversion efficiency of high heat storage batteries. According to this model, when ξ(t-Δt), h2(t-Δt), and P2(t-Δt) are input, the battery electrothermal conversion efficiency ξ(t) at the next time t can be predicted; h z (t) is the total conversion heat power during the thermoelectric conversion process of the high heat density solid thermal storage battery at time t; h3(t) is the heat power of the heat supply during the thermoelectric conversion process of the high heat density solid thermal storage battery at time t; Q S (t) is the residual heat capacity of the high heat density solid thermal storage battery at time t; Q(t1) is the heat capacity of the high heat density solid thermal storage battery at time t1; Q min P is the minimum heat storage capacity of the thermal storage battery; 2,max and P 2,min are the maximum and minimum electric power of the thermal-electric conversion model of the thermal storage battery; h 2,max and h 2,min are the maximum and minimum thermal powers converted into electrical energy by the thermal-electric conversion model of the thermal storage battery; h 3,max and h 3,min are the maximum and minimum heating powers of the thermal-electric conversion model of the thermal storage battery; h z,max and h z,min are the total maximum and minimum thermal powers of the thermal-electric conversion model of the thermal storage battery, respectively.

Images