CN113108632A

CN113108632A - Three-heat-source shell-and-tube heat exchanger capable of switching heat sources according to temperature

Info

Publication number: CN113108632A
Application number: CN202010566304.XA
Authority: CN
Inventors: 邱燕; 魏民; 王鑫; 张海静; 张井志; 刘延华; 郭亮; 刘昳娟; 鞠文杰; 王为帅; 孙卓新; 潘佳; 王者龙; 牛蔚然; 韩小岗; 王志梁; 其他发明人请求不公开姓名
Original assignee: State Grid Shandong Integrated Energy Service Co ltd; Shandong University; State Grid Shandong Electric Power Co Ltd
Current assignee: State Grid Shandong Integrated Energy Service Co ltd; Shandong University; State Grid Shandong Electric Power Co Ltd
Priority date: 2020-06-19
Filing date: 2020-06-19
Publication date: 2021-07-13
Anticipated expiration: 2040-06-19
Also published as: CN113108632B

Abstract

The invention provides a heat exchanger for switching heating according to temperature memory, wherein a first temperature sensor, a second temperature sensor and a third temperature sensor are respectively arranged in a left side pipe, a central pipe and a right side pipe, the temperature data of the first temperature sensor, the second temperature sensor and the third temperature sensor are stored in a database in real time, a one-dimensional deep convolution neural network is adopted to extract data characteristics, and pattern recognition is carried out, so that whether a first heat source, a third heat source and a second heat source are heated or not is controlled. Based on a theoretical method of machine learning and pattern recognition, the invention designs corresponding working states of the heat exchanger (whether the first heat source 91, the third heat source 93 and the second heat source 92 are heated or not), and trains a deep convolution neural network by using a large amount of temperature data, thereby controlling heat exchange and descaling of the heat exchanger.

Description

Three-heat-source shell-and-tube heat exchanger capable of switching heat sources according to temperature

Technical Field

The invention relates to a shell-and-tube heat exchanger, in particular to a shell-and-tube heat exchanger for intermittent vibration descaling.

Background

The shell-and-tube heat exchanger is widely applied to industries such as chemical industry, petroleum industry, refrigeration industry, nuclear energy industry and power industry, and due to the worldwide energy crisis, the demand of the heat exchanger in industrial production is more and more, and the quality requirement of the heat exchanger is higher and more. In recent decades, although compact heat exchangers (plate type, plate fin type, pressure welded plate type, etc.), heat pipe type heat exchangers, direct contact type heat exchangers, etc. have been rapidly developed, because the shell and tube type heat exchangers have high reliability and wide adaptability, they still occupy the domination of yield and usage, and according to relevant statistics, the usage of the shell and tube type heat exchangers in the current industrial devices still accounts for about 70% of the usage of all heat exchangers.

After the shell-and-tube heat exchanger is scaled, the heat exchanger is cleaned by adopting conventional modes of steam cleaning, back flushing and the like, and the production practice proves that the effect is not good. The end socket of the heat exchanger can only be disassembled, and a physical cleaning mode is adopted, but the mode is adopted for cleaning, so that the operation is complex, the consumed time is long, the investment of manpower and material resources is large, and great difficulty is brought to continuous industrial production.

The mode of passively strengthening heat exchange is to strictly prevent the fluid vibration induction in the heat exchanger from being changed into effective utilization of vibration, so that the convective heat transfer coefficient of the transmission element at low flow speed is greatly improved, dirt on the surface of the heat transfer element is restrained by vibration, the thermal resistance of the dirt is reduced, and the composite strengthened heat transfer is realized.

In application, it is found that continuous heating can cause the internal fluid to form stability, i.e. the fluid does not flow or has little fluidity, or the flow is stable, so that the vibration performance of the heat exchange tube is greatly weakened, thereby affecting the descaling of the heat exchange tube and the heating efficiency.

Current shell and tube heat exchangers include dual headers, one header evaporating and one header condensing, thereby forming a vibrating descaled heat pipe. Thereby improving the heat exchange efficiency of the heat pipe and reducing scaling. However, the heat pipe has insufficient uniformity of heat exchange, only one side is used for condensation, and the heat exchange amount is small, so that improvement is needed to develop a heat pipe system with a novel structure. There is therefore a need for improvements to the above-described heat exchangers.

In the prior application, a three-heat-source shell-and-tube heat exchanger has been developed, but the shell-and-tube heat exchanger is controlled according to the period, so that the vibration heat exchange effect is poor, and the intelligent degree is low. The present application therefore provides further improvements over the previous studies.

Disclosure of Invention

The invention provides an electric heating shell-and-tube heat exchanger with a novel structure, aiming at the defects of the shell-and-tube heat exchanger in the prior art. The heat exchanger can be based on a theoretical method of machine memory and pattern recognition, a corresponding operation mode is designed by utilizing temperature data in a heater real-time monitoring system according to different operation conditions of the heat exchanger, and a deep convolution neural network is trained by utilizing a large amount of temperature data, so that heat exchange part descaling is carried out, and the heat utilization effect and the descaling effect are improved. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

In order to achieve the purpose, the invention adopts the following technical scheme:

a heat exchanger capable of switching heating according to temperature memory comprises a shell, wherein tube plates are respectively arranged at two ends of the shell, a heat exchange component is arranged in the shell, the heat exchange component comprises a central tube, a left tube, a right tube and a tube group, the tube group comprises a left tube group and a right tube group, the left tube group is communicated with the left tube and the central tube, the right tube group is communicated with the right tube and the central tube, the central tube, the left tube, the central tube and the right tube are respectively provided with a first heat source, a second heat source and a third heat source, each tube group comprises a plurality of circular arc-shaped annular tubes, the end parts of the adjacent annular tubes are communicated, so that the plurality of annular tubes form a serial structure, and the end parts of the annular tubes form free ends of the annular tubes; the central tube comprises a first tube orifice and a second tube orifice, the first tube orifice is connected with the inlet of the left tube group, the second tube orifice is connected with the inlet of the right tube group, the outlet of the left tube group is connected with the left tube, and the outlet of the right tube group is connected with the right tube; the first pipe orifice and the second pipe orifice are arranged on the same side of the central pipe, and the left pipe group and the right pipe group are in mirror symmetry along the plane of the axis of the central pipe; a left return pipe is arranged between the left side pipe and the central pipe, and a right return pipe is arranged between the right side pipe and the central pipe; the system is characterized in that a first temperature sensor, a second temperature sensor and a third temperature sensor are respectively arranged in the left side pipe, the central pipe and the right side pipe and used for detecting the temperature in the left side pipe, the central pipe and the right side pipe, the first temperature sensor, the second temperature sensor and the third temperature sensor are in data connection with a controller, the temperature data of the first temperature sensor, the second temperature sensor and the third temperature sensor are stored in a database in real time, a one-dimensional deep convolutional neural network is adopted to extract data characteristics, pattern recognition is carried out, and therefore whether the first heat source, the third heat source and the second heat source are heated or not is controlled.

Preferably, the heat source is an electric heater.

The invention has the following advantages:

1. according to different operation conditions of the heat exchanger, the temperature data in the heat exchanger real-time monitoring system is utilized to design a corresponding operation mode (whether the first heat source 91, the third heat source 93 and the second heat source 92 are in a heating state or not), and a large amount of temperature data is used for training a deep convolution neural network, so that the heat exchange and descaling of the heat exchanger are controlled. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

2. The 3 heat sources of the invention heat alternately in a period, and can realize frequent vibration of the elastic coil, thereby realizing good descaling and heating effects and ensuring that the heating power is basically the same in time.

3. The invention increases the heating power of the coil pipe periodically and continuously and reduces the heating power, so that the heated fluid can generate the volume which is continuously in a changing state after being heated, and the free end of the coil pipe is induced to generate vibration, thereby strengthening heat transfer.

4. The heating efficiency can be further improved by the arrangement of the pipe diameter and the interval distribution of the pipe groups in the length direction.

5. The invention optimizes the optimal relationship of the parameters of the shell-and-tube heat exchanger through a large amount of experiments and numerical simulation, thereby realizing the optimal heating efficiency.

6. The invention designs a triangular layout diagram of a multi-heat exchange component with a novel structure, optimizes the structural parameters of the layout, and can further improve the heating efficiency through the layout.

Description of the drawings:

fig. 1 is a schematic view of a housing structure.

Fig. 2 is a top view of a heat exchange member of the present invention.

Fig. 3 is a front view of the heat exchange member of the present invention.

Fig. 4 is a front view of another embodiment of a heat exchange member of the present invention.

Fig. 5 is a dimensional structure schematic diagram of the heat exchange component of the invention.

Fig. 6 is a schematic layout of the heat exchange member of the present invention in a circular cross-section heater.

In the figure: 1. tube group, left tube group 11,

right tube group

12, 21, left tube, 22, right tube, 3, free end, 4, free end, 5, free end, 6, free end, 7, annular tube, 8, central tube, 91-93, heat source, 10 first tube orifice, 13 second tube orifice, left return tube 14, right return tube 15, front tube plate 16, support 17, support 18, rear tube plate 19,

shell

20, 21, shell inlet connecting tube, 22, shell outlet connecting tube, heat exchange component 23

Detailed Description

A shell-and-tube heat exchanger, as shown in fig. 1, comprises a shell 20, a heat exchange component 23, a shell-side inlet connecting pipe 21 and a shell-side outlet connecting pipe 22; the heat exchange component 23 is arranged in the shell 20 and fixedly connected to the front tube plate 16 and the rear tube plate 19; the shell side inlet connecting pipe 21 and the shell side outlet connecting pipe 22 are both arranged on the shell 20; fluid enters from the shell side inlet connecting pipe 21, exchanges heat through the heat exchange part and exits from the shell side outlet connecting pipe 22.

Preferably, the heat exchange member extends in a horizontal direction. The heat exchanger is arranged in the horizontal direction.

Fig. 2 shows a top view of a heat exchange unit 23, which, as shown in fig. 2, comprises a central tube 8, a left tube 21, a right tube 22 and a tube bank 1, the tube set 1 comprises a left tube set 11 and a right tube set 12, the left tube set 11 being in communication with a left side tube 21 and a central tube 8, the right tube set 12 being in communication with a right side tube 22 and the central tube 8, so that the central tube 8, the left side tube 21, the right side tube 22 and the tube group 1 form a closed circulation of heating fluid, the left side tube 21 and/or the central tube 8 and/or the right side tube 22 are filled with phase-change fluid, the left side tube 21, the central tube 8 and the right side tube 22 are respectively provided with a first heat source 91, a second heat source 92 and a third heat source 93, each tube group 1 comprises a plurality of circular arc-shaped annular tubes 7, the end parts of the adjacent annular tubes 7 are communicated, the plurality of annular tubes 7 form a serial structure, and the end parts of the annular tubes 7 form free ends 3-6 of the annular tubes; the central tube comprises a first tube orifice 10 and a second tube orifice 13, the first tube orifice 10 is connected with the inlet of the left tube group 11, the second tube orifice 13 is connected with the inlet of the right tube group 12, the outlet of the left tube group 11 is connected with the left tube 21, and the outlet of the right tube group 12 is connected with the right tube 22; the first orifice 10 and the second orifice 13 are arranged on the same side of the central tube 8. The left tube group and the right tube group are in mirror symmetry along the plane of the axis of the central tube.

The ends of the two ends of the center tube 8, the left tube 21 and the right tube 22 are disposed in the openings of the front and rear tube plates 16, 19 for fixation. The first orifice 10 and the second orifice 13 are located on the upper side of the central tube 8.

Preferably, a left return pipe 14 is arranged between the left pipe 21 and the central pipe 8, and a right return pipe 14 is arranged between the right pipe 22 and the central pipe 8. Preferably, the return pipe is arranged at the end of the central pipe. Both ends of the central tube are preferred.

Preferably, the fluid is a phase-change fluid, a vapor-liquid phase-change fluid, the first heat source 91, the second heat source 92 and the third heat source 93 are in data connection with a controller, and the controller controls the first heat source 91, the second heat source 92 and the third heat source 93 to heat.

The fluid is heated and evaporated in the central tube 8, flows to the left and

right headers

21 and 22 along the annular tube bundle, and is heated to expand in volume, so that steam is formed, and the volume of the steam is far larger than that of water, so that the formed steam can flow in the coil in a rapid impact manner. Because of volume expansion and steam flow, the free end of the annular tube can be induced to vibrate, the vibration is transmitted to the surrounding heat exchange fluid by the free end of the heat exchange tube in the vibration process, and the fluid can also generate disturbance, so that the surrounding heat exchange fluid forms disturbance flow, a boundary layer is damaged, and the purpose of enhancing heat transfer is realized. The fluid is condensed and released heat in the left and right side pipes and then flows back to the central pipe through the return pipe. Conversely, the fluid may be heated in the left and right pipes, condensed in the central pipe, and returned to the left and right pipes through the return pipe to be circulated.

According to the invention, the prior art is improved, and the condensation collecting pipe and the pipe groups are respectively arranged into two pipes which are distributed on the left side and the right side, so that the pipe groups distributed on the left side and the right side can perform vibration heat exchange descaling, the heat exchange vibration area is enlarged, the vibration can be more uniform, the heat exchange effect is more uniform, the heat exchange area is increased, and the heat exchange and descaling effects are enhanced.

The 3 heat sources are alternately heated in a period, and the periodic frequent vibration of the elastic coil can be realized, so that good descaling and heating effects are realized, and the heating power is basically the same in time.

Preferably, the annular pipes of the left pipe group are distributed by taking the axis of the left pipe as the center of a circle, and the annular pipes of the right pipe group are distributed by taking the axis of the right pipe as the center of a circle. The left side pipe and the right side pipe are arranged as circle centers, so that the distribution of the annular pipes can be better ensured, and the vibration and the heating are uniform.

Preferably, the tube group is plural.

Preferably, the center tube 8, the left tube 21, and the right tube 22 are provided along the longitudinal direction.

Preferably, the left tube group 21 and the right tube group 22 are staggered in the longitudinal direction, as shown in fig. 3. Through the staggered distribution, can make to vibrate heat transfer and scale removal on different length for the vibration is more even, strengthens heat transfer and scale removal effect.

Preferably, the tube group 2 is provided in plural (for example, the same side (left side or right side)) along the length direction of the center tube 8, and the tube diameter of the tube group 2 (for example, the same side (left side or right side)) becomes larger along the flow direction of the fluid in the shell side.

Preferably, the pipe diameter of the annular pipe of the pipe group (for example, the same side (left side or right side)) is increased along the flowing direction of the fluid in the shell side.

The pipe diameter range through the heat exchange tube increases, can guarantee that shell side fluid outlet position fully carries out the heat transfer, forms the heat transfer effect like the adverse current, further strengthens the heat transfer effect moreover for whole vibration effect is even, and the heat transfer effect increases, further improves heat transfer effect and scale removal effect. Experiments show that better heat exchange effect and descaling effect can be achieved by adopting the structural design.

Preferably, the tube group on the same side (left side or right side) is provided in plural along the length direction of the center tube 8, and the distance between the adjacent tube groups on the same side (left side or right side) becomes smaller along the flow direction of the fluid in the shell side.

Preferably, the spacing between the tube banks on the same side (left or right) in the direction of fluid flow in the shell side is increased by a decreasing amount.

The interval amplitude through the heat exchange tube increases, can guarantee that shell side fluid outlet position fully carries out the heat transfer, forms the heat transfer effect like the adverse current, further strengthens the heat transfer effect moreover for the whole vibration effect is even, and the heat transfer effect increases, further improves heat transfer effect and scale removal effect. Experiments show that better heat exchange effect and descaling effect can be achieved by adopting the structural design.

In tests it was found that the pipe diameters, distances and pipe diameters of the left side pipe 21, the right side pipe 22, the central pipe 8 and the pipe diameters of the ring pipes can have an influence on the heat exchange efficiency and the uniformity. If the distance between the collector is too big, then heat exchange efficiency is too poor, and the distance between the ring shape pipe is too little, then the ring shape pipe distributes too closely, also can influence heat exchange efficiency, and the pipe diameter size of collector and heat exchange tube influences the volume of the liquid or the steam that holds, then can exert an influence to the vibration of free end to influence the heat transfer. Therefore, the pipe diameters and distances of the left pipe 21, the right pipe 22, the central pipe 8 and the pipe diameters of the ring pipes have a certain relationship.

The invention provides an optimal size relation summarized by numerical simulation and test data of a plurality of heat pipes with different sizes. Starting from the maximum heat exchange amount in the heat exchange effect, nearly 200 forms are calculated. The dimensional relationship is as follows:

the distance between the center of the central tube 8 and the center of the left tube 21 is equal to the distance between the center of the central tube 8 and the center of the right tube 21, L, the tube diameter of the left tube 21, the tube diameter of the central tube 8, and the radius of the right tube 22 are R, the radius of the axis of the innermost annular tube in the annular tubes is R1, and the radius of the axis of the outermost annular tube is R2, so that the following requirements are met:

R1/R2 ═ a × Ln (R/L) + b; where a, b are parameters and Ln is a logarithmic function, where 0.6212< a <0.6216, 1.300< b < 1.301; preferably, a is 0.6214 and b is 1.3005.

Preferably, 35< R <61 mm; 114< L <190 mm; 69< R1<121mm, 119< R2<201 mm.

Preferably, the number of annular tubes of the tube set is 3-5, preferably 3 or 4.

Preferably, 0.55< R1/R2< 0.62; 0.3< R/L < 0.33.

Preferably, 0.583< R1/R2< 0.615; 0.315< R/L < 0.332.

Preferably, the radius of the annular tube is preferably 10-40 mm; preferably 15 to 35mm, more preferably 20 to 30 mm.

Preferably, the centers of the left tube 21, the right tube 22 and the center tube 8 are on a straight line.

Preferably, the arc between the ends of the free ends 3, 4 around the centre axis of the left tube is 95-130 degrees, preferably 120 degrees. The same applies to the curvature of the free ends 5, 6 and the free ends 3, 4. Through the design of the preferable included angle, the vibration of the free end is optimal, and therefore the heating efficiency is optimal.

Preferably, the heat exchange component can be used as an immersed heat exchange assembly, immersed in a fluid to heat the fluid, for example, the heat exchange component can be used as an air radiator heating assembly, and can also be used as a water heater heating assembly.

The heating power of the first, second and third heat sources is preferably 1000-.

Preferably, the box body has a circular cross section, and is provided with a plurality of heat exchange components, wherein one heat exchange component is arranged at the center of the circular cross section (the center pipe is arranged at the center of the circle) and the other heat exchange components are distributed around the center of the circular cross section.

Preferably, the tube bundle of the tube bank 1 is an elastic tube bundle.

The heat exchange coefficient can be further improved by arranging the tube bundle of the tube group 1 with an elastic tube bundle.

Further preferably, the heat source is an electric heating rod.

The number of the pipe groups 1 is multiple, and the plurality of pipe groups 1 are in a parallel structure.

The heat exchanger shown in fig. 6 has a circular cross-sectional housing in which the plurality of heat exchange members are disposed. Preferably, the number of the heat exchange components is three, the center of the central tube of each heat exchange component is located at the midpoint of an inscribed regular triangle of the circular cross section, the connecting lines of the centers of the central tubes form the regular triangle, one heat exchange component is arranged at the upper part of each central tube, two heat exchange components are arranged at the lower part of each central tube, and the connecting lines formed by the left side tube, the right side tube and the centers of the central tubes of the heat exchange components are of a parallel structure. Through such setting, can make and to fully reach vibrations and heat transfer purpose in can making the heater, improve the heat transfer effect.

Learn through numerical simulation and experiment, heat transfer part's size and circular cross-section's diameter have very big influence to the heat transfer effect, heat transfer part size too big can lead to adjacent interval too little, the space that the centre formed is too big, middle heating effect is not good, the heating is inhomogeneous, on the same way, heat transfer part size undersize can lead to adjacent interval too big, leads to whole heating effect not good. Therefore, the invention obtains the optimal size relation through a large amount of numerical simulation and experimental research.

The distance between the centers of the left side pipe and the right side pipe is L1, the side length of the inscribed regular triangle is L2, the radius of the axis of the innermost annular pipe in the annular pipes is R1, and the radius of the axis of the outermost annular pipe is R2, so that the following requirements are met:

10*(L1/L2)＝d*(10*R1/R2)-e*(10*R1/R2)²-f; wherein d, e, f are parameters,

44.102<d<44.110,3.715<e<3.782,127.385<f<127.395；

more preferably, d is 44.107, e is 3.718, f is 127.39;

with 720< L2<1130mm preferred. Preferably 0.58< R1/R2< 0.62.

Further preferred is 0.30< L1/L2< 0.4.

Through the layout of the three heat exchange component structure optimization, the whole heat exchange effect can reach the best heat exchange effect.

It has been found in research and practice that a constant and stable heat source results in a fluid-forming stability of the internal heat exchange components, i.e. the fluid is not flowing or is less fluid, or the flow is stable, resulting in a significantly reduced vibration performance of the tube bank 1, which affects the efficiency of the descaling and heating of the tube bank 1. Therefore, the following improvements are required for the heat pipe.

In the prior application of the inventor, a periodic heat exchange mode is provided, and the vibration of the annular tube is continuously promoted through the periodic heat exchange mode, so that the heat exchange efficiency and the descaling effect are improved. However, adjusting the vibration of the tube bundle with a fixed periodic variation can lead to hysteresis and too long or too short a period. Therefore, the invention improves the previous application and intelligently controls the vibration, so that the fluid in the fluid can realize frequent vibration, and good descaling and heat exchange effects are realized.

Aiming at the defects in the technology researched in the prior art, the invention provides a novel heat exchanger capable of intelligently controlling vibration. The heat exchanger can improve the heat exchange efficiency, thereby realizing good descaling and heat exchange effects.

Self-regulation vibration based on pressure

Preferably, a first pressure sensor, a second pressure sensor and a third pressure sensor are respectively arranged in the left side pipe 21, the central pipe 8 and the right side pipe 22 and are used for detecting the pressure in the left side pipe, the central pipe and the right side pipe, the first pressure sensor, the second pressure sensor and the third pressure sensor are in data connection with the controller,

the pressure data of the first pressure sensor, the second pressure sensor and the third pressure sensor are stored in a database in real time, data features are extracted by adopting a one-dimensional deep convolution neural network, and pattern recognition is carried out, so that whether the first heat source 91, the third heat source 93 and the second heat source 92 are heated or not is controlled.

The pressure-based autonomous adjustment vibration pattern recognition includes the steps of:

1. preparing data: and reexamining and checking the pressure data in the database, correcting missing data, invalid data and inconsistent data, and ensuring the correctness and logical consistency of the data.

2. Generating a data set: the prepared data is divided into training set/training set labels, detection set/detection set labels.

3. Network training: inputting the training set data into a convolution neural network, continuously performing convolution and pooling to obtain a characteristic vector, and sending the characteristic vector into a full-connection network. And obtaining a network error by calculating the output of the network and a training set label, and continuously correcting the network weight, the bias, the convolution coefficient and the pooling coefficient by using an error back propagation algorithm to enable the error to meet the set precision requirement, thereby finishing the network training.

4. Network detection: and inputting the detection set data into the trained network, and outputting a detection result label.

5. The heat exchanger operates: and controlling whether the first and

third heat sources

91 and 93 and the second heat source 92 are heated or not according to the detection result label so as to carry out heat exchange and descaling.

The invention provides a novel system for intelligently controlling vibration descaling of a heat exchange device, which is based on a theoretical method of machine learning and pattern recognition, utilizes pressure data with time correlation in a centralized heat exchanger real-time monitoring system according to different operating conditions of the heat exchanger, designs corresponding working states of the heat exchanger (whether a first heat source 91, a third heat source 93 and a second heat source 92 are in a heating state), trains a deep convolutional neural network by using a large amount of pressure data, and thus controls the heat exchange and descaling of the heat exchanger.

Preferably, the data preparation step specifically includes the following processing:

1) processing missing data: missing values in the database may occur due to a failure of the network transmission. For the missing data value, adopting an estimation method and replacing the missing value with the sample mean value;

2) processing invalid data: the pressure data in the database may have invalid values, such as negative values or values exceeding a theoretical maximum value, due to a failure of the sensor, and these values are deleted from the database;

3) processing inconsistent data: the inconsistent data is checked by means of an integrity constraint mechanism of the database management system, and then corrected by referring to corresponding data values in the database. Preferably, in the heat exchanger, the pressure of the heated tube of the first and

third heat sources

91, 93 and the second heat source 92 is necessarily higher than the pressure of the non-heated tube, and if the pressure of the heated tube in the database is lower than the pressure of the non-heated tube, a user error prompt may be given by means of a check constraint mechanism in the integrity constraint of the database management system, and the user replaces the pressure data value of the inconsistent data with the estimated data or the corresponding critical pressure data value according to the error prompt.

Preferably, the step of generating a data set comprises the steps of:

1) generating training set data and labels: and reading the pressure data values of the corresponding working conditions from the database according to different operating conditions of the heat collecting device, and generating training set data and working condition labels under various working condition states. Preferably, in a specific application, the operation condition is divided into a state labeled as 1, in which the first and

third heat sources

91 and 93 are heated and the second heat source 92 is not heated, a state labeled as 2, in which the first and

third heat sources

91 and 93 are not heated and the second heat source 92 is heated. Automatically generating working condition labels by a program according to different working conditions;

preferably, the data includes data indicating that the evaporation of the fluid within the internal heat exchange component is substantially saturated or stable under different operating conditions. The working condition comprises at least one of valve opening, heat exchange fluid temperature and the like.

2) Generating detection set data and labels: and reading the pressure data values of the corresponding working conditions from the database according to different operating conditions of the heat exchanger, and generating detection set data and working condition labels under various working condition states. The working condition labels are the same as the working condition labels of the training set and are automatically generated by a program according to the running working conditions.

Preferably, it is determined whether or not the evaporation of the fluid inside the left tube, the right tube (heated by the first heat source, the third heat source), or the center tube (heated by the second heat source) is saturated or stable (reaches or exceeds a certain pressure). For example, the left and right tubes are not saturated or stable, labeled 11 for saturation or stability, labeled 12 for saturation or stability, the center tube is not saturated or stable, labeled 21 for saturation or stability, and labeled 22.

The network training comprises the following specific steps:

1) reading a group of training set data d, wherein the size of the training set data d is [ Mx 1 xN ], M represents the size of a training batch, and 1 xN represents one-dimensional training data;

2) and performing a first convolution operation on the read training data to obtain a feature map t. Initializing coefficients of a convolution kernel g, and setting the size of g as [ P × 1 × Q ], wherein P represents the number of convolution kernels, [1 × Q ] represents the size of the convolution kernels, the obtained convolution result is t ═ Σ (d × g), and the size of a feature map is [ M × 1 × N × Q ];

3) and performing maximum pooling operation on the feature map t obtained by the convolution operation to obtain a feature map z. Initializing a pooling coefficient, wherein the given pooling step length is p, the size of a pooling window is k, the size of a finally obtained feature map z is [ Mx1 x (N/p). times.Q ], and the data dimensionality is reduced in a pooling process;

4) repeating the steps 2) -3), repeatedly performing convolution and pooling operation to obtain a feature vector x, and finishing the feature extraction process of the convolutional neural network;

5) initializing a weight matrix w and an offset b of the full-connection network, sending the extracted eigenvector x into the full-connection network, and calculating with the weight matrix w and the offset b to obtain a network output y ═ sigma (wxx + b);

6) subtracting the training set label l from the output y obtained by the network to obtain a network error e which is y-l, carrying out derivation on the network error, and sequentially correcting the weight w, the bias b, the pooling coefficients of each layer and the convolution coefficients of each layer of the fully-connected network by utilizing the derivative back propagation;

7) and repeating the process until the network error e meets the precision requirement, finishing the network training process, and generating a convolutional neural network model.

When the first heat source and the third heat source heat and the second heat source does not heat, the data are measured by the first pressure sensor and the third pressure sensor. Preferably, the average of the first and third pressures is used. When the first heat source, the third heat source do not heat and the second heat source heats, the data is the data measured by the second pressure sensor.

The network detection steps are as follows:

1) loading the trained convolutional neural network model, wherein the convolutional kernel coefficient, the pooling coefficient, the network weight w and the bias b of the convolutional neural network are trained;

2) and inputting the detection data set into the trained convolutional neural network, and outputting a detection result. The type of run can be determined, for example, based on the output tag. For example, 1 represents a heating state of the first and

third heat sources

91 and 93 and a non-heating state of the second heat source 92, 2 represents a non-heating state of the first and

third heat sources

91 and 93 and a heating state of the second heat source 92, and so on.

The invention provides a new method for controlling heat exchange of a heat exchanger, which makes full use of online monitoring data of the heat exchanger, and has the advantages of high detection speed and low cost.

The invention organically integrates the data processing technology, machine learning and pattern recognition theory, and can improve the accuracy of the operation of the heat exchanger.

The working process of the specific convolutional neural network is as follows:

1) inputting a group of training set data d, wherein the size of the training set data d is [ M multiplied by 1 multiplied by N ], M represents the size of the training batch, and 1 multiplied by N represents one-dimensional training data;

3) and performing maximum pooling operation on the feature map t obtained by the convolution operation to obtain a feature map z. Initializing a pooling coefficient, setting a pooling step length as p, setting a pooling window size as k, and reducing data dimensionality in a pooling process, wherein the size of a finally obtained feature map z is [ MX 1X (N/p) XQ ];

4) repeating the steps 2) -3), and repeatedly performing convolution and pooling operation to obtain a feature vector;

the heat exchanger comprises the following operation steps:

1) when the first heat source and the third heat source heat and the second heat source does not heat, the result label output in the step 4 shows that the evaporation of the fluid inside the left tube and the right tube reaches saturation or stability, the controller controls the first heat source and the third heat source to stop heating, and the second heat source heats;

2) and when the first heat source and the third heat source stop heating and the second heat source heats, the result label output by the step 4 shows that the evaporation of the fluid in the central tube reaches saturation or stability, the controller controls the first heat source and the third heat source to heat, and the second heat source stops heating.

Repeating step 1) or 2).

By means of pattern recognition of pressure detected by the pressure sensing element, the heat exchanger scale removal method can design a corresponding operation mode by utilizing pressure data in a heat exchanger real-time monitoring system based on a theoretical method of machine memory and pattern recognition according to different operation conditions of the heat exchanger, and train a deep convolution neural network by utilizing a large amount of pressure data, so that scale removal of heat exchange parts is carried out, and the heat utilization effect and the scale removal effect are improved. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

The invention can be based on a theoretical method of machine memory and mode recognition, and the pressure detected by the pressure sensing element can ensure that the evaporation of the fluid in the left side pipe, the right side pipe or the central pipe is basically saturated and the volume of the internal fluid is not changed greatly under the condition of meeting a certain pressure. Therefore, new fluid is started to perform alternate heat exchange by detecting the pressure change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

The invention can be based on the theoretical method of machine memory and pattern recognition, so that the detection and judgment results are more accurate.

Through the pressure that pressure perception element detected, can satisfy under certain pressure condition, the evaporation of the inside fluid of left side pipe, right side pipe or center tube has reached saturation basically, and the volume of inside fluid also changes little basically, and under this kind of condition, inside fluid is relatively stable, and the tube bank vibratility variation at this moment consequently needs to be adjusted, changes heat exchange component, makes the fluid flow towards different directions. Therefore, new heat sources are started to perform alternate heat exchange by detecting the pressure change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

Independently adjusting vibration based on temperature

Preferably, a first temperature sensor, a second temperature sensor and a third temperature sensor are respectively arranged in the left side tube 21, the central tube 8 and the right side tube 22 and used for detecting the temperature in the left side tube, the central tube and the right side tube, the first temperature sensor, the second temperature sensor and the third temperature sensor are in data connection with the controller, the temperature data of the first temperature sensor, the second temperature sensor and the third temperature sensor are stored in a database in real time, a one-dimensional deep convolution neural network is adopted to extract data characteristics and perform pattern recognition, and therefore whether the first heat source 91, the third heat source 93 and the second heat source 92 are heated or not is controlled.

The temperature-based self-adjusting vibration pattern recognition comprises the following steps:

1. preparing data: and reexamining and checking the temperature data in the database, correcting missing data, invalid data and inconsistent data, and ensuring the correctness and logical consistency of the data.

5. The heat exchanger operates: and controlling whether the first and

third heat sources

The invention provides a novel system for intelligently controlling vibration descaling of a heat exchange device, which is based on a theoretical method of machine learning and pattern recognition, utilizes temperature data with time correlation in a centralized heat exchanger real-time monitoring system according to different operating conditions of the heat exchanger, designs corresponding working states of the heat exchanger (whether a first heat source 91, a third heat source 93 and a second heat source 92 are in a heating state), trains a deep convolutional neural network by using a large amount of temperature data, and thereby controls the heat exchange descaling of the heat exchanger.

2) processing invalid data: the temperature data in the database may have invalid values, such as negative values or values exceeding a theoretical maximum value, due to a failure of the sensor, and these values are deleted from the database;

3) processing inconsistent data: the inconsistent data is checked by means of an integrity constraint mechanism of the database management system, and then corrected by referring to corresponding data values in the database. Preferably, in the heat exchanger, the temperature of the tubes heated by the first, third and

second heat sources

91, 93 and 92 is always higher than that of the non-heated tubes, and if the temperature of the heated tubes in the database is lower than that of the non-heated tubes, a user error prompt can be given by means of a check constraint mechanism in the integrity constraint of the database management system, and the user replaces the temperature data value of the inconsistent data with the estimated data or the corresponding critical temperature data value according to the error prompt.

Preferably, the step of generating a data set comprises the steps of:

1) generating training set data and labels: and reading the temperature data values of the corresponding working conditions from the database according to different operating conditions of the heat collecting device, and generating training set data and working condition labels under various working condition states. Preferably, in a specific application, the operation condition is divided into a state labeled as 1, in which the first and

third heat sources

2) Generating detection set data and labels: and reading the temperature data values of the corresponding working conditions from the database according to different operating conditions of the heat exchanger, and generating detection set data and working condition labels under various working condition states. The working condition labels are the same as the working condition labels of the training set and are automatically generated by a program according to the running working conditions.

Preferably, it is determined whether or not the evaporation of the fluid inside the left tube, the right tube (heated by the first heat source and the third heat source), or the center tube (heated by the second heat source) is saturated or stable (reaches or exceeds a predetermined temperature). For example, the left and right tubes are not saturated or stable, labeled 11 for saturation or stability, labeled 12 for saturation or stability, the center tube is not saturated or stable, labeled 21 for saturation or stability, and labeled 22.

The network training comprises the following specific steps:

When the first heat source and the third heat source heat and the second heat source does not heat, the data are measured by the first temperature sensor and the third temperature sensor. Preferably, the average of the first and third temperatures is used. When the first heat source, the third heat source do not heat and the second heat source heats, the data is the data measured by the second temperature sensor.

The network detection steps are as follows:

third heat sources

91 and 93 and a heating state of the second heat source 92, and so on.

The working process of the specific convolutional neural network is as follows:

the heat exchanger comprises the following operation steps:

Repeating step 1) or 2).

Through the mode recognition of the temperature detected by the temperature sensing element, the invention can design a corresponding operation mode by utilizing the temperature data in the heat exchanger real-time monitoring system according to different operation conditions of the heat exchanger based on the theoretical method of machine memory and mode recognition, and train the deep convolution neural network by utilizing a large amount of temperature data, thereby descaling the heat exchange part and improving the heat utilization effect and the descaling effect. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

The invention can be based on a theoretical method of machine memory and mode recognition, and the temperature detected by the temperature sensing element can ensure that the evaporation of the fluid in the left side pipe, the right side pipe or the central pipe is basically saturated and the volume of the internal fluid is not changed greatly under the condition of meeting a certain temperature, and under the condition, the internal fluid is relatively stable, the vibration of the tube bundle is poor, so that the adjustment is needed, the heat exchange component is changed, and the fluid flows towards different directions. Therefore, new fluid is started to perform alternate heat exchange by detecting the temperature change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

The temperature that detects through temperature perception element can satisfy under certain temperature condition, the evaporation of the inside fluid of left side pipe, right side pipe or center tube has basically reached saturation, and the volume of inside fluid also changes little basically, and under this kind of condition, inside fluid is relatively stable, and the tube bank vibratility variation at this moment, consequently needs adjust, changes heat exchange component, makes the fluid flow towards different directions. Therefore, new heat sources are started to perform alternate heat exchange by detecting the temperature changes in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

Thirdly, automatically adjusting vibration based on liquid level

Preferably, a first liquid level sensor, a second liquid level sensor and a third liquid level sensor are respectively arranged in the left side pipe 21, the central pipe 8 and the right side pipe 22 and used for detecting liquid levels in the left side pipe, the right side pipe and the central pipe, the first liquid level sensor, the second liquid level sensor and the third liquid level sensor are in data connection with the controller, liquid level data of the first liquid level sensor, the second liquid level sensor and the third liquid level sensor are stored in a database in real time, a one-dimensional deep convolution neural network is adopted for extracting data characteristics and performing mode identification, and therefore whether the first heat source 91, the third heat source 93 and the second heat source 92 are heated or not is controlled.

The liquid level-based autonomous regulation vibration pattern recognition comprises the following steps:

1. preparing data: and (4) reexamining and verifying the liquid level data in the database, and correcting missing data, invalid data and inconsistent data to ensure the correctness and logical consistency of the data.

5. The heat exchanger operates: and controlling whether the first and

third heat sources

The invention provides a novel system for intelligently controlling vibration descaling of a heat exchange device, which is based on a theoretical method of machine learning and pattern recognition, utilizes liquid level data with time correlation in a centralized heat exchanger real-time monitoring system according to different operating conditions of the heat exchanger, designs corresponding working states of the heat exchanger (whether a first heat source 91, a third heat source 93 and a second heat source 92 are in a heating state or not), and trains a deep convolutional neural network by using a large amount of liquid level data so as to control heat exchange and descaling of the heat exchanger.

2) processing invalid data: due to a failure of the sensor, the liquid level data in the database has invalid values, such as negative values or values exceeding a theoretical maximum value, and the values are deleted from the database;

3) processing inconsistent data: the inconsistent data is checked by means of an integrity constraint mechanism of the database management system, and then corrected by referring to corresponding data values in the database. Preferably, in the heat exchanger, the liquid level of the pipe heated by the first and

third heat sources

91 and 93 and the second heat source 92 is necessarily lower than the liquid level of the non-heated pipe, and if the liquid level of the heated pipe in the database is higher than the liquid level of the non-heated pipe, a user error prompt can be given by means of a check constraint mechanism in the integrity constraint of the database management system, and the user replaces the liquid level data value of the inconsistent data with the estimated data or the corresponding critical liquid level data value according to the error prompt.

Preferably, the step of generating a data set comprises the steps of:

1) generating training set data and labels: and reading the liquid level data values of the corresponding working conditions from the database according to different operating conditions of the heat collecting device, and generating training set data and working condition labels under various working condition states. Preferably, in a specific application, the operation condition is divided into a state labeled as 1, in which the first and

third heat sources

preferably, the data includes data indicating that the evaporation of the fluid within the internal heat exchange component is substantially saturated or stable under different operating conditions. The operating condition comprises at least one of valve opening, heat exchange fluid level and the like.

2) Generating detection set data and labels: and reading the liquid level data values of the corresponding working conditions from the database according to different operating conditions of the heat exchanger, and generating detection set data and working condition labels under various working condition states. The working condition labels are the same as the working condition labels of the training set and are automatically generated by a program according to the running working conditions.

Preferably, it is determined whether or not the evaporation of the fluid inside the left tube, the right tube (heated by the first heat source, the third heat source), or the center tube (heated by the second heat source) is saturated or stabilized (reaches or falls below a certain level). For example, the left and right tubes are not saturated or stable, labeled 11 for saturation or stability, labeled 12 for saturation or stability, the center tube is not saturated or stable, labeled 21 for saturation or stability, and labeled 22.

The network training comprises the following specific steps:

5) initializing a weight matrix w and a bias b of the fully-connected network, sending the extracted eigenvector x into the fully-connected network, and calculating with the weight matrix w and the bias b to obtain a network output y-sigma (wxx + b);

When the first heat source and the third heat source heat and the second heat source does not heat, the data are measured by the first liquid level sensor and the third liquid level sensor. Preferably, the average of the first and third liquid levels is used. When the first heat source and the third heat source do not heat and the second heat source heats, the data is measured by the second liquid level sensor.

The network detection steps are as follows:

third heat sources

91 and 93 and a heating state of the second heat source 92, and so on.

The working process of the specific convolutional neural network is as follows:

the heat exchanger comprises the following operation steps:

Repeating step 1) or 2).

Through mode recognition of the liquid level detected by the liquid level sensing element, the invention can design a corresponding operation mode by utilizing liquid level data in a real-time monitoring system of the heat exchanger according to different operation conditions of the heat exchanger based on a theoretical method of machine memory and mode recognition, and train a deep convolution neural network by utilizing a large amount of liquid level data, thereby descaling heat exchange parts and improving the heat utilization effect and the descaling effect. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

The invention can be based on a theoretical method of machine memory and mode recognition, and the liquid level detected by the liquid level sensing element can ensure that the evaporation of the fluid in the left side pipe, the right side pipe or the central pipe is basically saturated and the volume of the internal fluid is not changed greatly under the condition of meeting a certain liquid level. Therefore, new fluid is started to perform alternate heat exchange by detecting the liquid level change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

Through the liquid level that liquid level perception element detected, can satisfy under certain liquid level condition, the evaporation of the inside fluid of left side pipe, right side pipe or center tube has reached saturation basically, and the volume of inside fluid also changes little basically, and under this kind of condition, inside fluid is relatively stable, and the tube bank vibratility variation at this moment is consequently poor, consequently need adjust, changes heat exchange component, makes the fluid flow towards different directions. Therefore, a new heat source is started to perform alternate heat exchange by detecting the liquid level change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

Fourthly, automatically adjusting vibration based on speed

Preferably, a speed sensing element is arranged inside the free end of the tube bundle and used for detecting the flow speed of fluid in the free end of the tube bundle, the speed sensing element is in data connection with the controller, speed data of the speed sensor is stored in a database in real time, a one-dimensional deep convolution neural network is adopted to extract data characteristics and perform pattern recognition, and therefore whether the first heat source 91, the third heat source 93 and the second heat source 92 are heated or not is controlled.

The speed-based autonomous adjustment vibration pattern recognition comprises the following steps:

1. preparing data: and (4) rechecking and checking the speed data in the database, and correcting missing data, invalid data and inconsistent data to ensure the correctness and logical consistency of the data.

5. The heat exchanger operates: and controlling whether the first and

third heat sources

The invention provides a novel system for intelligently controlling vibration descaling of a heat exchange device, which is based on a theoretical method of machine learning and pattern recognition, utilizes speed data with time correlation in a centralized heat exchanger real-time monitoring system according to different operating conditions of the heat exchanger, designs corresponding working states of the heat exchanger (whether a first heat source 91, a third heat source 93 and a second heat source 92 are in a heating state or not), and trains a deep convolutional neural network by using a large amount of speed data so as to control heat exchange and descaling of the heat exchanger.

2) processing invalid data: the speed data in the database is invalid due to a failure of the sensor, such as negative values or exceeding a theoretical maximum value, and the values are deleted from the database;

3) processing inconsistent data: the inconsistent data is checked by means of an integrity constraint mechanism of the database management system, and then corrected by referring to corresponding data values in the database. Preferably, in the heat exchanger, the first, third and

second heat sources

91, 93 and 92 heat pipes at a speed higher than the speed of the non-heated pipes, and if the speed of the heated pipes in the database is lower than the speed of the non-heated pipes, a user error prompt may be given by means of a check constraint mechanism in the integrity constraint of the database management system, and the user replaces the speed data value of the non-uniform data with the estimated data or the corresponding critical speed data value according to the error prompt.

Preferably, the step of generating a data set comprises the steps of:

1) generating training set data and labels: and reading speed data values of corresponding working conditions from the database according to different operating conditions of the heat collecting device, and generating training set data and working condition labels under various working condition states. Preferably, in a specific application, the operation condition is divided into a state labeled as 1, in which the first and

third heat sources

preferably, the data includes data indicating that the evaporation of the fluid within the internal heat exchange component is substantially saturated or stable under different operating conditions. The operating condition comprises at least one of valve opening, heat exchange fluid speed and the like.

2) Generating detection set data and labels: and reading speed data values corresponding to working conditions from the database according to different operating conditions of the heat exchanger, and generating detection set data and working condition labels under various working condition states. The working condition labels are the same as the working condition labels of the training set and are automatically generated by a program according to the running working conditions.

Preferably, it is determined whether or not the evaporation of the fluid inside the left tube, the right tube (heated by the first heat source and the third heat source), or the center tube (heated by the second heat source) is saturated or stable (reaches or exceeds a certain speed). For example, the left and right tubes are not saturated or stable, labeled 11 for saturation or stability, labeled 12 for saturation or stability, the center tube is not saturated or stable, labeled 21 for saturation or stability, and labeled 22.

The network training comprises the following specific steps:

When the first heat source and the third heat source heat and the second heat source does not heat, the positive direction data is adopted as the data. When the first heat source and the third heat source do not heat and the second heat source heats, the data adopts negative direction data.

The network detection steps are as follows:

third heat sources

91 and 93 and a heating state of the second heat source 92, and so on.

The working process of the specific convolutional neural network is as follows:

the heat exchanger comprises the following operation steps:

Repeating step 1) or 2).

Through the mode recognition of the speed detected by the speed sensing element, the invention can design a corresponding operation mode by utilizing the speed data in the real-time monitoring system of the heat exchanger according to different operation conditions of the heat exchanger based on the theoretical method of machine memory and mode recognition, and train the deep convolution neural network by using a large amount of speed data, thereby descaling the heat exchange part and improving the heat utilization effect and the descaling effect. The shell-and-tube heat exchanger can realize the periodic frequent vibration of the heat exchange tube, and improves the heating efficiency, thereby realizing good descaling and heating effects.

The invention can be based on a theoretical method of machine memory and mode recognition, and the speed detected by the speed sensing element can ensure that the evaporation of the fluid in the left side pipe, the right side pipe or the central pipe is basically saturated and the volume of the internal fluid is not changed greatly under the condition of meeting a certain speed, so that the internal fluid is relatively stable and the vibration of the pipe bundle is poor, and therefore, the adjustment is needed to change the heat exchange component to ensure that the fluid flows towards different directions. Therefore, new fluid is started to perform alternate heat exchange by detecting the speed change in the left side pipe, the right side pipe and the central pipe, and the heat exchange effect and the descaling effect are improved.

The speed detected by the speed sensing element can basically reach saturation of the evaporation of the internal fluid under the condition of meeting a certain speed (such as the highest upper limit), so that stable flow is formed, and the speed of the internal fluid basically does not change greatly. And a new heat source is started to perform alternate heat exchange by detecting the speed change, so that the heat exchange effect and the descaling effect are improved.

Preferably, the speed sensing element is disposed at the free end. Through setting up at the free end, can perceive the speed change of free end to realize better control and regulation.

Preferably, the heat source is an electric heater.

Preferably, the axes of the left tube, the right tube and the middle tube are connected in a straight line or on a plane.

Preferably, the pipe diameters of the left side pipe and the right side pipe are smaller than the pipe diameter of the middle pipe. The pipe diameter of the middle pipe is preferably 1.4-1.5 times of the pipe diameter of the left side pipe and the right side pipe. Through the pipe diameter setting of left side pipe, right side pipe and intermediate pipe, can guarantee that the fluid carries out the phase transition and keeps the same or close transmission speed at left side pipe, right side pipe and intermediate pipe to guarantee the homogeneity of conducting heat.

Preferably, the connection position of the coil pipe at the left channel box is lower than the connection position of the central pipe and the coil pipe. This ensures that steam can rapidly pass up into the central tube. Similarly, the connecting position of the coil pipe at the right channel box is lower than the connecting position of the central pipe and the coil pipe

Although the present invention has been described with reference to the preferred embodiments, it is not limited thereto. Various changes and modifications may be effected therein by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A heat exchanger capable of switching heating according to temperature memory comprises a shell, wherein tube plates are respectively arranged at two ends of the shell, a heat exchange component is arranged in the shell, the heat exchange component comprises a central tube, a left tube, a right tube and a tube group, the tube group comprises a left tube group and a right tube group, the left tube group is communicated with the left tube and the central tube, the right tube group is communicated with the right tube and the central tube, the central tube, the left tube, the central tube and the right tube are respectively provided with a first heat source, a second heat source and a third heat source, each tube group comprises a plurality of circular arc-shaped annular tubes, the end parts of the adjacent annular tubes are communicated, so that the plurality of annular tubes form a serial structure, and the end parts of the annular tubes form free ends of the annular tubes; the central tube comprises a first tube orifice and a second tube orifice, the first tube orifice is connected with the inlet of the left tube group, the second tube orifice is connected with the inlet of the right tube group, the outlet of the left tube group is connected with the left tube, and the outlet of the right tube group is connected with the right tube; the first pipe orifice and the second pipe orifice are arranged on the same side of the central pipe, and the left pipe group and the right pipe group are in mirror symmetry along the plane of the axis of the central pipe; a left return pipe is arranged between the left side pipe and the central pipe, and a right return pipe is arranged between the right side pipe and the central pipe; the system is characterized in that a first temperature sensor, a second temperature sensor and a third temperature sensor are respectively arranged in the left side pipe, the central pipe and the right side pipe and used for detecting the temperature in the left side pipe, the central pipe and the right side pipe, the first temperature sensor, the second temperature sensor and the third temperature sensor are in data connection with a controller, the temperature data of the first temperature sensor, the second temperature sensor and the third temperature sensor are stored in a database in real time, a one-dimensional deep convolutional neural network is adopted to extract data characteristics, pattern recognition is carried out, and therefore whether the first heat source, the third heat source and the second heat source are heated or not is controlled.

2. The heat exchanger of claim 1, wherein the temperature memory based switching heating method comprises the steps of:

1) preparing data; 2) generating a data set; 3) network training; 4) network detection; 5) the heat exchanger is operated.

3. A shell-and-tube heat exchanger is characterized by comprising a plurality of circular arc-shaped annular tubes, wherein the end parts of the adjacent annular tubes are communicated, so that the annular tubes form a series structure.