CN114894346A - Phase-change heat storage monitoring system and method based on neural network - Google Patents
Phase-change heat storage monitoring system and method based on neural network Download PDFInfo
- Publication number
- CN114894346A CN114894346A CN202210451674.8A CN202210451674A CN114894346A CN 114894346 A CN114894346 A CN 114894346A CN 202210451674 A CN202210451674 A CN 202210451674A CN 114894346 A CN114894346 A CN 114894346A
- Authority
- CN
- China
- Prior art keywords
- heat storage
- phase
- phase change
- temperature
- heat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K17/00—Measuring quantity of heat
- G01K17/06—Measuring quantity of heat conveyed by flowing media, e.g. in heating systems e.g. the quantity of heat in a transporting medium, delivered to or consumed in an expenditure device
- G01K17/08—Measuring quantity of heat conveyed by flowing media, e.g. in heating systems e.g. the quantity of heat in a transporting medium, delivered to or consumed in an expenditure device based upon measurement of temperature difference or of a temperature
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/14—Thermal energy storage
Landscapes
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Combustion & Propulsion (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention discloses a phase change heat storage monitoring system based on a neural network, which comprises a phase change heat storage device, wherein the phase change heat storage device comprises a sealed shell and a phase change material positioned in the shell, and a heat exchange pipeline for heat exchange is arranged in the phase change material; the phase change material is a solid-liquid phase change material; the system also comprises a detection system and a heat storage prediction system; the detection system comprises a pressure sensor and a plurality of temperature sensors; the shell is filled with gas which does not react with the phase-change material chemically, and the pressure sensor is used for measuring the pressure of the gas; the temperature sensor is used for measuring the temperature of the phase change material at different positions in the shell; the heat storage prediction system is constructed based on a neural network and trained by adopting experimental data, and the heat storage prediction system inputs detection values of the pressure sensors and the temperature sensors and outputs the current heat storage amount of the phase-change heat storage device. The invention has high precision and stable operation, not only monitors the temperature distribution of the phase-change material in the phase-change heat storage device in real time, but also can feed back the current heat storage capacity in real time.
Description
Technical Field
The invention relates to a phase-change heat storage monitoring system and method, in particular to a phase-change heat storage monitoring system and method based on a neural network.
Background
Currently, the share of renewable energy in power supply systems is increasing year by year. With the continuous expansion of the renewable power generation scale represented by wind power and photovoltaic, the randomness and the fluctuation caused by the renewable power generation scale cause the reduction of the flexibility of a power grid and the insufficient peak regulation capability, so that the phenomenon of 'wind abandoning and light abandoning' is increased, and finally, the economical efficiency of the operation of the wind power generation and the photovoltaic power generation is seriously influenced; in order to solve the above problems, energy storage technologies must be vigorously developed. The energy storage technology represented by phase change heat storage can convert redundant electric energy into heat energy to be stored in the valley power period, and the heat energy is taken out in the peak power period to meet daily production requirements, so that the electric energy consumption in the peak power period is reduced, the effects of peak clipping and valley filling are achieved, and the operation of a power grid is stabilized. In order to make the phase change heat storage device serve the energy system better, the operation state of the phase change heat storage device needs to be precisely controlled. Due to the particularity of the phase change heat storage process, it is difficult to obtain an accurate heat storage state through simple temperature measurement. At present, the mainstream measurement method is to assemble a heat meter in a system, measure the heat input into the phase change heat storage device and the heat output from the phase change heat storage device respectively, and use the difference between the two as the heat storage capacity of the current phase change heat storage device. However, the high-precision ultrasonic calorimeter is expensive, the property of fluid in the system and the state of the pipe wall of the pipeline have great influence on the measurement precision, and meanwhile, the measurement mode ignores heat leakage factors in the heat charging and discharging process of the phase change heat storage device, so that the measurement deviation of the calorimeter is increasingly increased after the calorimeter is operated for a long time. Therefore, an efficient, stable and economical measurement system and an implementation method are needed to realize accurate measurement of the phase-change heat storage amount.
Disclosure of Invention
The invention provides a phase change heat storage monitoring system and method based on a neural network for solving the technical problems in the prior art.
The technical scheme adopted by the invention for solving the technical problems in the prior art is as follows: a phase change heat storage monitoring system based on a neural network comprises a phase change heat storage device, wherein the phase change heat storage device comprises a sealed shell and a phase change material positioned in the shell, and a heat exchange pipeline for heat exchange is arranged in the phase change material; the phase change material is a solid-liquid phase change material; the system also comprises a detection system and a heat storage prediction system; the detection system comprises a pressure sensor and a plurality of temperature sensors; the shell is filled with gas which does not react with the phase-change material chemically, and the pressure sensor is used for measuring the pressure of the gas; the temperature sensor is used for measuring the temperature of the phase change material at different positions in the shell; the heat storage prediction system is constructed based on a neural network and trained by adopting experimental data, and the heat storage prediction system inputs detection values of the pressure sensors and the temperature sensors and outputs the current heat storage amount of the phase-change heat storage device.
Further, the solid volume of the phase change material is 75% -80% of the internal volume of the shell.
Further, the housing is filled with nitrogen or an inert gas.
Furthermore, the temperature sensors are arranged in multiple layers from top to bottom, each layer of temperature sensors is N, and the distance between every two adjacent layers is equal.
Furthermore, the number of the temperature sensors on each layer is five, one temperature sensor is located at the center point of the cross section of the shell, and the other four temperature sensors are located on a straight line which passes through the center point and is perpendicular to each other, and the center distance of the temperature sensors is the same as that of the temperature sensors located at the center point.
Further, the distance between the temperature sensor at the central point and the other four temperature sensors is equal to the distance between the other four temperature sensors and the inner side surface of the shell.
Further, the shell is square; four temperature sensors located in the same layer outside the center point are located on the center line of the cross section.
The invention also provides a phase change heat storage monitoring method based on the neural network of the phase change heat storage monitoring system based on the neural network, which comprises the following steps:
measuring heat energy absorbed and/or released by a phase change heat storage device in a heat storage process by using a calibrated calorimeter, simultaneously acquiring detection values of a pressure sensor and each temperature sensor to obtain a plurality of groups of experimental data, and manufacturing a training set and a verification set by using the experimental data;
step two, taking the detection values of the pressure sensors and the temperature sensors as model inputs, taking the current heat storage amount as model output, and constructing a neural network model of the heat storage amount prediction system;
step three: training and verifying the neural network model by using a training set and a verification set;
step four: and inputting the detection values of the pressure sensor and the temperature sensor into the trained neural network model, and outputting the current heat storage amount of the phase-change heat storage device in real time by the heat storage amount prediction system.
Further, the first step comprises the following sub-steps:
step A1, measuring heat energy input and output by the heat exchange pipeline by using a calorimeter, and carrying out a complete heat charging test on the phase change heat storage device to obtain the maximum heat storage capacity Q of the phase change heat storage device, wherein Q is Q a -Q loss (ii) a Wherein Q is a Is the count value, Q, of the calorimeter loss Releasing the heat leakage quantity to the outside for the phase change heat storage device; wherein Q loss H × a × Δ T × T; in the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T is the difference between the surface temperature of the shell and the temperature of the outside air; t is the time length of corresponding heat charging and discharging;
step A2, charging heat to the phase change heat storage device to make the heat storage quantity of the phase change heat storage device reach the maximum heat storage quantity Q, and collecting the detection values of each temperature sensor and each pressure sensor at the moment;
step A3, releasing heat to the phase change heat storage device, every time the reading variation of the calorimeter is Q 1 Collecting the detection values of a primary temperature sensor and a primary pressure sensor, and simultaneously calculating and storing the heat storage capacity of the current phase-change heat storage device until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 1 When the temperature is higher than the set temperature, the heat storage amount corresponding to the phase change heat storage device is Q-nQ 1 -Q loss-d Wherein Q is loss-d ＝h×A×△T d ×t d (ii) a In the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T d The difference value between the surface temperature of the shell and the temperature of the outside air in the heat release process is obtained; t is t d The cumulative duration of the corresponding heat release;
step A4, the phase change heat storage device is charged with heat, and each time the reading variation of the calorimeter is Q 2 Collecting the detection values of a primary temperature sensor and a primary pressure sensor, and simultaneously calculating and storing the heat storage capacity of the current phase-change heat storage device until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 2 When the temperature is higher than the predetermined temperature, the heat storage amount corresponding to the phase change heat storage device is nQ 2 -Q loss-c Wherein Q is loss-c ＝h×A×△T c ×t c (ii) a In the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T c The difference value between the surface temperature of the shell and the temperature of the outside air in the heat charging process; t is t c The accumulated time length of the corresponding heat charging;
and step A5, repeating the steps A3 to A4 for M times to obtain multiple groups of experimental data, making one part of the experimental data into a training set, and making the other part of the experimental data into a verification set.
Further, Q 1 ＝Q 2 ＝1％×Q。
The invention has the advantages and positive effects that: the phase change heat storage monitoring system based on the neural network adopts the temperature sensor and the pressure sensor which are relatively low in cost, the arrangement mode is simple, and the measurement result is stable and reliable. The technical scheme is high in precision and stable in operation, the temperature distribution of the phase-change material in the phase-change heat storage device can be monitored in real time, and the current heat storage capacity can be fed back in real time according to the temperature distribution. The neural network model can be used as a standardized product for mass production, an early-stage producer can construct and train the neural network model of the heat storage prediction system according to the neural network-based phase-change heat storage monitoring method, the trained neural network model can be built in the standardized product monitoring system, a user can directly use the model for heat storage monitoring without retraining, and the cost is saved.
Drawings
Fig. 1 is a schematic structural diagram of a phase-change heat storage monitoring system based on a neural network.
Fig. 2 is a schematic sectional view taken along line a-a of fig. 1.
In the figure: 1. a pressure sensor; 2. a heat exchange coil; 3. a temperature sensor; 4. a housing.
Detailed Description
For further understanding of the contents, features and effects of the present invention, the following embodiments are enumerated in conjunction with the accompanying drawings, and the following detailed description is given:
referring to fig. 1 and 2, a phase change heat storage monitoring system based on a neural network includes a phase change heat storage device, the phase change heat storage device includes a sealed housing 4 and a phase change material located in the housing 4, and a heat exchange pipeline for heat exchange is arranged in the phase change material; the phase change material is a solid-liquid phase change material; the system also comprises a detection system and a heat storage prediction system; the detection system comprises a pressure sensor 1 and a plurality of temperature sensors 3; the shell 4 is filled with gas which does not react with the phase-change material chemically, and the pressure sensor 1 is used for measuring the pressure of the gas; the temperature sensor 3 is used for measuring the temperature of the phase change material at different positions in the shell 4; the heat storage prediction system is constructed based on a neural network and trained by adopting experimental data, and the heat storage prediction system inputs detection values of the pressure sensor 1 and the temperature sensors 3 and outputs the current heat storage amount of the phase change heat storage device.
Preferably, the solid volume of the phase change material may be 75-80% of the internal volume of the housing 4.
Preferably, the housing 4 may be filled with nitrogen or an inert gas.
Preferably, the temperature sensors 3 may be arranged in multiple layers from top to bottom, each layer of temperature sensors 3 may be N, and the distance between two adjacent layers may be equal.
Preferably, each layer of the temperature sensors 3 may be five, one temperature sensor 3 may be located at a central point of the cross section of the housing 4, and the remaining four temperature sensors 3 may be located on a straight line passing through the central point and perpendicular to each other, and have the same center distance as the temperature sensor 3 located at the central point.
Preferably, the distance between the temperature sensor 3 at the central point and the other four temperature sensors 3 can be equal to the distance between the other four sensors and the inner side surface of the shell.
Preferably, the housing 4 may be square; the four temperature sensors 3 located in the same layer outside the center point may be located on the center line of the cross section.
Preferably, the housing 4 may comprise an inner layer and an outer layer; the outer layer can be made of metal, and the inner layer can be an insulating layer.
The neural network may be any suitable neural network known in the art, such as a CNN neural network. The temperature sensor, the pressure sensor, the phase change heat storage device and the like can adopt applicable components and devices in the prior art, the phase change material can adopt the phase change material in the prior art,
the invention also provides a phase change heat storage monitoring method based on the neural network of the phase change heat storage monitoring system based on the neural network, which comprises the following steps:
firstly, measuring heat energy absorbed and/or released by the phase change heat storage device in the heat storage process by using a calibrated calorimeter, simultaneously acquiring detection values of the pressure sensor 1 and each temperature sensor 3 to obtain a plurality of groups of experimental data, and manufacturing a training set and a verification set by using the experimental data.
And step two, taking the detection values of the pressure sensor 1 and the temperature sensors 3 as model inputs, taking the current heat storage amount as model output, and constructing a neural network model of the heat storage amount prediction system.
Step three: and training and verifying the neural network model by using the training set and the verification set.
Step four: and inputting the detection values of the pressure sensor and the temperature sensor into the trained neural network model, and outputting the current heat storage amount of the phase-change heat storage device in real time by the heat storage amount prediction system.
Preferably, the first step may comprise the following sub-steps:
step A1, heat energy input and output by the heat exchange pipeline can be measured by a calorimeter, and the phase change heat storage device can be completely charged with heat onceThe maximum heat storage capacity Q of the phase change heat storage device is obtained through testing, and Q is equal to Q a -Q loss (ii) a Wherein Q is a Is the count value, Q, of the calorimeter loss Releasing the heat leakage quantity to the outside for the phase change heat storage device; wherein Q loss H × a × Δ T × T; in the formula, h is the surface convection heat transfer coefficient of the shell 4; a is the outer surface area of the shell 4; Δ T is the difference between the surface temperature of the case 4 and the outside air temperature; t is the time period corresponding to the heat charge and discharge.
In step a2, the phase change heat storage device is charged with heat to reach the maximum heat storage amount Q, and the detection values of the temperature sensor 3 and the pressure sensor 1 at this time are collected.
Step A3, releasing heat to the phase change heat storage device, wherein the reading change quantity of the calorimeter is Q 1 Then, the detection values of the primary temperature sensor 3 and the pressure sensor 1 are collected, and the heat storage capacity of the current phase change heat storage device is calculated and stored at the same time until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 1 When the temperature is higher than the set temperature, the heat storage amount corresponding to the phase change heat storage device is Q-nQ 1 -Q loss-d ，Q loss-d The phase change heat storage device releases heat leakage to the outside in the heat release process.
Wherein Q loss-d ＝h×A×△T d ×t d (ii) a In the formula, h is the surface convection heat transfer coefficient of the shell 4; a is the outer surface area of the shell 4; delta T d Is the difference between the surface temperature of the shell 4 and the outside air temperature in the heat release process; t is t d The cumulative duration of the corresponding heat release;
step A4, charging the phase change heat storage device, wherein the reading change quantity of each calorimeter is Q 2 Then, the detection values of the primary temperature sensor 3 and the pressure sensor 1 are collected, and the heat storage capacity of the current phase change heat storage device is calculated and stored at the same time until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 2 When the temperature is higher than the predetermined temperature, the heat storage amount corresponding to the phase change heat storage device is nQ 2 -Q loss-c ，Q loss-c The heat leakage quantity released to the outside by the phase change heat storage device in the heat charging process is realized.
Wherein Q loss-c ＝h×A×△T c ×t c (ii) a In the formula, hThe surface convection heat transfer coefficient of the shell 4; a is the external surface area of the shell 4; delta T c Is the difference between the surface temperature of the shell 4 and the outside air temperature in the heat charging process; t is t c Is the cumulative time corresponding to the heat charge.
Step a5, repeating steps A3 to a4 for M times to obtain multiple sets of experimental data, wherein a part of the experimental data can be made into a training set, and the other part of the experimental data can be made into a verification set.
The heat data of the input and output phase-change heat storage device is collected through the calorimeter, the fixed signal variable quantity of the heat storage device can be used as a collection time node when heat is released, timing can be started when the heat storage quantity reaches the maximum heat storage quantity Q, and timing can be started when the heat storage quantity is zero when heat is filled. Taking the fixed reading variation of the calorimeter as the acquisition time node, the time node can be recorded, for example, from the time node T, each time the reading of the calorimeter changes by a certain amount, i.e., each time the reading of the calorimeter increases or decreases by a certain fixed value n Starting timing when the calorimeter is from T n Time to T n+1 When the reading number is increased or decreased to a set value at the moment, T n 、T n+1 May be the time node for collecting data, and the time period between two time nodes is T ═ T n+1 -T n 。
Therefore, the time for inputting and outputting certain heat quantity from the calorimeter can be calculated, and the heat leakage quantity can be estimated through a formula, so that the heat storage quantity of each time node of the phase change heat storage device can be estimated.
Preferably, Q 1 、Q 2 The values of (a) can be as follows: q 1 ＝Q 2 ＝1％×Q。
The structure, the working process and the working principle of the present invention are further explained by a preferred embodiment of the present invention as follows:
a phase change heat storage monitoring system based on a neural network is shown in figure 1 and comprises a shell 4, a heat exchange coil 2, a temperature sensor 3 and a pressure sensor 1. The inner layer of the shell 4 close to the phase-change material is made of heat insulation materials, such as nano aerogel, glass wool and the like, and is used for reducing the outward heat dissipation capacity of the phase-change heat storage material; the outer layer is wrapped by metal pressure-bearing material, such as 304 stainless steel, and the like, and is used for ensuring the pressure-bearing effect of the shell 4.
The phase change heat storage device is characterized in that a phase change heat storage material is filled in a shell 4, and the volume of the phase change heat storage material is 75% -80% of the internal volume of the shell 4; the phase change heat storage material can be selected according to the actual heat supply temperature requirement, and the phase change temperature range can be selected as follows: 50-300 ℃; the inside residual volume of casing 4 fills in nitrogen gas, plays and prevents that inside components and parts from oxidizing, improves effects such as pressure sensor 1 measurement stability.
The temperature sensors 3 are uniformly distributed in each layer of phase change material along the axial direction, 5 temperature sensors 3 are simultaneously distributed in each layer, and the temperature sensors 3 are arranged at equal intervals along the symmetry axis; pressure sensor 1 installs at 4 tops of casing to be linked together with the cavity is inside, and pressure sensor 1 chooses the gaseous type pressure sensor 1 of high accuracy, small range for use, recommends the range: 0 to 1 MPa.
The invention selects temperature and pressure as measuring variables, and the theoretical basis is as follows: because the phase change heat storage material heating in-process is heated unevenly, can appear the temperature layering in the cavity vertical direction after producing the density difference, even the different position temperature in same layer also has certain difference, consequently temperature sensor 3 arranges respectively in order to obtain the temperature numerical value that corresponds different region along axial and horizontal direction in the phase change heat storage material respectively. The temperature of the phase-change heat storage material changes, the density of the phase-change heat storage material inevitably changes, and finally the volume of the phase-change heat storage material changes, especially the volume changes obviously due to solid-liquid conversion in the phase-change process; therefore, when the heated volume of the phase-change material is increased, nitrogen on the upper part of the cavity is extruded, so that the numerical value of the pressure sensor 1 is correspondingly changed; finally, the current state of the phase-change heat storage material can be predicted according to the changes of the temperature and the pressure, and the current heat storage amount can be obtained.
A phase change heat storage monitoring method based on a neural network comprises the following steps:
step 1: preparing a high-precision calorimeter, calibrating before experiment, mounting the calorimeter at the front end of the heat exchange coil 2, and carrying out one-time complete heat charging test on the phase change heat storage device to obtain the phase change heat storage deviceThe maximum heat storage capacity of the device is Q, and Q is Q a -Q loss (ii) a Wherein Q is a For measuring the value, Q, of the calorimeter loss For heat leakage to the outside. Wherein Q loss H × a × Δ T × T; in the formula, h is the surface convection heat transfer coefficient of the shell 4; a is the outer surface area of the shell 4; Δ T is the difference between the surface temperature of the case 4 and the outside air temperature; t is the corresponding duration of the heat charge.
Step 2: when the phase change heat storage device is full of heat Q, collecting the numerical values recorded by the current temperature sensor 3 and the current pressure sensor 1;
and step 3: releasing heat of the phase change heat storage device, wherein the released heat is Q d (measured by a calorimeter) taking into account the amount Q of heat leakage from the shell 4 to the outside during the heat release process loss-d Heat release quantity Q d ＝1％×Q-Q loss-d (ii) a After releasing heat at present, gather the numerical value of temperature sensor 3 and pressure sensor 1 record to record the heat accumulation volume that current phase change heat storage device corresponds, promptly: (N) d -1)％×Q，N d The initial value is 100, N after each subsequent action d The value minus 1;
and 4, step 4: repeating the third action for 99 times until the heat storage capacity in the phase change heat storage device is emptied to 0;
and 5: charging the phase change heat storage device with the charging quantity of Q c (measured by a calorimeter) considering the amount Q of heat leakage from the casing 4 to the outside during the heat charging process loss-c Heat charging quantity Q c ＝1％×Q+Q loss-c After the current heat charging, the numerical value recorded by the temperature sensor 3 and the pressure sensor 1 is collected, and the heat storage amount corresponding to the current phase change heat storage device is recorded, namely: (N) c +1)％×Q，N c The initial value is 0, N is obtained after each subsequent action c The value is increased by 1;
step 6: repeating the fifth action for 99 times until the heat storage quantity in the phase change heat storage device is charged to Q;
and 7: repeating the second step to the sixth step for 100 times, and collating the data acquired in the previous steps to form a sample set (A) i ，B i )；A i As data (i.e., temperature and pressure data per acquisition), B i Is a label (i.e. corresponding to the current phase-change heat storage capacity), is alwaysForming 20100 groups of samples; 80% of the samples are made into a training sample set, and 20% of the samples are made into a verification sample set.
And 8: and (3) taking the collected temperature and pressure values as model input and the current heat storage amount as model output to construct a neural network model of the heat storage amount of the phase-change heat storage device. In this example, the neural network model inputs layer 46 nodes, hidden layer 128 nodes, and outputs layer 1 nodes;
and step 9: each time a training sample is extracted from the training sample set and sent to the neural network for training, the actual output Y of the network is calculated i (ii) a According to the error D ═ B i -Y i (namely how much the predicted value is different from the actual value) adjusting the weight w of the neural network; the above process is repeated for each training sample until the error does not exceed the specified range for the entire sample set (the percentage error is less than 0.5% in this example). And verifying the trained model by adopting the verification samples in the verification sample set.
Step 10: the trained neural network is stored in a computer or a microcomputer, and the current heat storage capacity can be calculated in real time according to the temperature and pressure data fed back by the phase change heat storage device in real time.
The above-mentioned embodiments are only for illustrating the technical ideas and features of the present invention, and the purpose thereof is to enable those skilled in the art to understand the contents of the present invention and to carry out the same, and the present invention shall not be limited to the embodiments, i.e. the equivalent changes or modifications made within the spirit of the present invention shall fall within the scope of the present invention.
Claims (10)
1. A phase change heat storage monitoring system based on a neural network comprises a phase change heat storage device, wherein the phase change heat storage device comprises a sealed shell and a phase change material positioned in the shell, and a heat exchange pipeline for heat exchange is arranged in the phase change material; the phase change material is a solid-liquid phase change material; the system is characterized by also comprising a detection system and a heat storage prediction system; the detection system comprises a pressure sensor and a plurality of temperature sensors; the shell is filled with gas which does not react with the phase-change material chemically, and the pressure sensor is used for measuring the pressure of the gas; the temperature sensor is used for measuring the temperature of the phase change material at different positions in the shell; the heat storage prediction system is constructed based on a neural network and trained by adopting experimental data, and the heat storage prediction system inputs detection values of the pressure sensors and the temperature sensors and outputs the current heat storage amount of the phase-change heat storage device.
2. The neural network-based phase-change thermal storage monitoring system of claim 1, wherein the solid volume of the phase-change material is 75% to 80% of the internal volume of the housing.
3. The neural network-based phase-change thermal storage monitoring system of claim 1, wherein the housing is filled with nitrogen or an inert gas.
4. The neural network-based phase-change heat storage monitoring system as claimed in claim 1, wherein the temperature sensors are arranged in multiple layers from top to bottom, N temperature sensors are arranged in each layer, and the distance between two adjacent layers is equal.
5. The neural network-based phase-change thermal storage monitoring system as claimed in claim 4, wherein the number of the temperature sensors in each layer is five, one of the temperature sensors is located at a central point of the cross section of the shell, and the other four temperature sensors are located on mutually perpendicular straight lines passing through the central point and are at the same center distance as the temperature sensors located at the central point.
6. The neural network-based phase-change thermal storage monitoring system of claim 5, wherein the temperature sensor located at the central point is equidistant from the other four temperature sensors and the other four sensors are equidistant from the inner side of the housing.
7. The neural network-based phase-change thermal storage monitoring system of claim 6, wherein the housing is square; four temperature sensors located in the same layer outside the center point are located on the center line of the cross section.
8. A neural network-based phase-change stored heat monitoring method using the neural network-based phase-change stored heat monitoring system according to any one of claims 1 to 7, comprising the steps of:
measuring heat energy absorbed and/or released by a phase change heat storage device in a heat storage process by using a calibrated calorimeter, simultaneously acquiring detection values of a pressure sensor and each temperature sensor to obtain a plurality of groups of experimental data, and manufacturing a training set and a verification set by using the experimental data;
step two, taking the detection values of the pressure sensors and the temperature sensors as model inputs, taking the current heat storage amount as model output, and constructing a neural network model of the heat storage amount prediction system;
step three: training and verifying the neural network model by using a training set and a verification set;
step four: and inputting the detection values of the pressure sensor and the temperature sensor into the trained neural network model, and outputting the current heat storage amount of the phase-change heat storage device in real time by the heat storage amount prediction system.
9. The method for monitoring phase-change stored heat based on the neural network as claimed in claim 8, wherein the first step comprises the following substeps:
step A1, measuring heat energy input and output by the heat exchange pipeline by using a calorimeter, and carrying out a complete heat charging test on the phase change heat storage device to obtain the maximum heat storage capacity Q of the phase change heat storage device, wherein Q is Q a -Q loss (ii) a Wherein Q is a Is the count value, Q, of the calorimeter loss Releasing the heat leakage quantity to the outside for the phase change heat storage device; wherein Q loss H × a × Δ T × T; in the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T is the difference between the surface temperature of the shell and the temperature of the outside air; t is the time length of corresponding heat charging and discharging;
step A2, charging heat to the phase change heat storage device to make the heat storage quantity of the phase change heat storage device reach the maximum heat storage quantity Q, and collecting the detection values of each temperature sensor and each pressure sensor at the moment;
step A3, releasing heat to the phase change heat storage device, every time the reading variation of the calorimeter is Q 1 Collecting the detection values of a primary temperature sensor and a primary pressure sensor, and simultaneously calculating and storing the heat storage capacity of the current phase-change heat storage device until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 1 When the temperature is higher than the set temperature, the heat storage amount corresponding to the phase change heat storage device is Q-nQ 1 -Q loss-d Wherein Q is loss-d ＝h×A×△T d ×t d (ii) a In the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T d The difference value between the surface temperature of the shell and the temperature of the outside air in the heat release process is obtained; t is t d The cumulative duration of the corresponding heat release;
step A4, the phase change heat storage device is charged with heat, and each time the reading variation of the calorimeter is Q 2 Collecting the detection values of a primary temperature sensor and a primary pressure sensor, and simultaneously calculating and storing the heat storage capacity of the current phase-change heat storage device until the reading variation of the calorimeter is zero; when the reading variation of the calorimeter is nQ 2 When the temperature is higher than the predetermined temperature, the heat storage amount corresponding to the phase change heat storage device is nQ 2 -Q loss-c Wherein Q is loss-c ＝h×A×△T c ×t c (ii) a In the formula, h is the surface convection heat transfer coefficient of the shell; a is the external surface area of the shell; delta T c The difference value between the surface temperature of the shell and the temperature of the outside air in the heat charging process; t is t c The accumulated time length of the corresponding heat charging;
and step A5, repeating the steps A3 to A4 for M times to obtain multiple groups of experimental data, making one part of the experimental data into a training set, and making the other part of the experimental data into a verification set.
10. The neural network-based phase-change stored heat monitoring method of claim 9, wherein Q is Q 1 ＝Q 2 ＝1％×Q。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210451674.8A CN114894346B (en) | 2022-04-27 | 2022-04-27 | Phase-change heat storage monitoring system and method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210451674.8A CN114894346B (en) | 2022-04-27 | 2022-04-27 | Phase-change heat storage monitoring system and method based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114894346A true CN114894346A (en) | 2022-08-12 |
| CN114894346B CN114894346B (en) | 2023-01-17 |
Family
ID=82718996
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210451674.8A Active CN114894346B (en) | 2022-04-27 | 2022-04-27 | Phase-change heat storage monitoring system and method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114894346B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103743972A (en) * | 2013-12-25 | 2014-04-23 | 青海中控太阳能发电有限公司 | Fault diagnosis method for tower type solar energy heat power generation system |
| JP2014152947A (en) * | 2013-02-05 | 2014-08-25 | Taisei Corp | Heat storage amount calculation method and heat storage amount calculation device |
| JP5635220B1 (en) * | 2013-10-18 | 2014-12-03 | 中国電力株式会社 | Heat storage amount prediction device, heat storage amount prediction method and program |
| CN106382668A (en) * | 2016-08-29 | 2017-02-08 | 东北大学 | System and method capable of realizing heating by combining electric heat storage boiler with solar energy |
| CN107389227A (en) * | 2017-08-09 | 2017-11-24 | 国家电网公司 | The measure device and its measuring method of a kind of phase-changing energy storage material residue amount of stored heat |
| JP2020159575A (en) * | 2019-03-25 | 2020-10-01 | いすゞ自動車株式会社 | Control system of heat accumulator and control method therefor |
-
2022
- 2022-04-27 CN CN202210451674.8A patent/CN114894346B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014152947A (en) * | 2013-02-05 | 2014-08-25 | Taisei Corp | Heat storage amount calculation method and heat storage amount calculation device |
| JP5635220B1 (en) * | 2013-10-18 | 2014-12-03 | 中国電力株式会社 | Heat storage amount prediction device, heat storage amount prediction method and program |
| CN103743972A (en) * | 2013-12-25 | 2014-04-23 | 青海中控太阳能发电有限公司 | Fault diagnosis method for tower type solar energy heat power generation system |
| CN106382668A (en) * | 2016-08-29 | 2017-02-08 | 东北大学 | System and method capable of realizing heating by combining electric heat storage boiler with solar energy |
| CN107389227A (en) * | 2017-08-09 | 2017-11-24 | 国家电网公司 | The measure device and its measuring method of a kind of phase-changing energy storage material residue amount of stored heat |
| JP2020159575A (en) * | 2019-03-25 | 2020-10-01 | いすゞ自動車株式会社 | Control system of heat accumulator and control method therefor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114894346B (en) | 2023-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Experimental and numerical study on the performance of a new high-temperature packed-bed thermal energy storage system with macroencapsulation of molten salt phase change material | |
| CN106154171B (en) | The method for establishing damping function in battery DC | |
| CN103675016B (en) | The test macro of a kind of release type tubular reactor phase transformation heat-obtaining and temperature control effect | |
| CN102121910B (en) | Performance test device for micro heat exchanger | |
| CN1982859A (en) | Electric-cable core temperature on-line monitoring system | |
| CN108627766B (en) | Real-time measurement method for internal temperature of battery core in battery module and battery pack | |
| CN103293013B (en) | A kind of cooling tower thermal performance appraisal procedure and system | |
| CN101013765A (en) | Method for real-time evaluating internal-external temperature difference of nickel-hydrogen electrokinetic cell | |
| CN101113963A (en) | Method and device for measuring liquid thermal conductivity factor | |
| Wang et al. | Heat generation measurement and thermal management with phase change material based on heat flux for high specific energy power battery | |
| CN114894346B (en) | Phase-change heat storage monitoring system and method based on neural network | |
| CN107179141A (en) | A kind of oil immersed type power transformation equipment temperature measuring device and calibration method with on-line calibration function | |
| Young et al. | Performance characteristics of a thermosyphon solar domestic hot water system | |
| CN114005552A (en) | Heat pipe reactor integrated solid reactor core structure capable of easily measuring reactor core temperature | |
| He et al. | Optimization design and performance investigation on the cascaded packed-bed thermal energy storage system with spherical capsules | |
| CN105954680B (en) | Device and method for testing heat generation power of power battery | |
| CN117387978A (en) | Liquid cooling plate performance test and data processing method | |
| Earnshaw et al. | The effective thermal conductivity of a bed of 1.2-mm-diam lithium zirconate spheres in helium | |
| CN109492322B (en) | Prediction method for spontaneous combustion position of coal body in coal storage silo | |
| CN204594914U (en) | Groove type solar heat collecting pipe heat waste loses test macro | |
| Mao et al. | Numerical and experimental investigation on heat transfer performance of a solar single storage tank | |
| CN205049262U (en) | Alternating temperature compensation calorimeter assembly of low excessive enthalpy of critical temperature liquid mixing | |
| CN104219302B (en) | Heating pipe network heat statistics decomposes acquiring and transmission system and method | |
| CN2932310Y (en) | Integrated circuit, measuring device convenient for self-calibration | |
| CN114062945A (en) | Method for detecting expansion force of lithium battery |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |