Thermal storage model prediction model construction and prediction method, device, medium and equipment

By acquiring and processing pressure difference signal samples, a thermal storage mode prediction model is constructed, which solves the problem of low prediction accuracy of thermal storage mode in the existing technology, achieving higher prediction accuracy and meeting the needs of single-well mining and irrigation technology.

CN116362393BActive Publication Date: 2024-11-08CHINA THREE GORGES CORPORATION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310294433.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-03-21
Publication Date
2024-11-08
Estimated Expiration
2043-03-21

AI Technical Summary

Technical Problem

The prediction accuracy of the existing thermal storage mode is low and it is difficult to meet the requirements of single-well mining and irrigation technology, mainly due to the uncertainty of the geological environment.

Method used

By obtaining the pressure difference signal sample set, feature extraction is performed to obtain multi-scale entropy and entropy increase rate. These features are used to train the preset prediction model, and the thermal storage mode prediction model is constructed and optimized to improve prediction accuracy.

Benefits of technology

It improves the accuracy of thermal storage mode prediction, can more accurately reflect the environmental changes of thermal storage, and meets the requirements of single-well mining and irrigation technology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116362393B_ABST
    Figure CN116362393B_ABST
Patent Text Reader

Abstract

The present invention discloses the construction of a prediction model for a heat storage mode and a prediction method, device, medium and equipment. The method comprises: obtaining a sample set of pressure difference signals, wherein the pressure difference signal sample set comprises pressure difference signal samples corresponding to the bottom of a deep well pump of a test well of various heat storage modes during water pumping, performing feature extraction processing on the pressure difference signal samples in the sample set to obtain multi-scale entropy, and calculating an entropy increase rate based on the multi-scale entropy, training a preset prediction model using a training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set and the heat storage mode of the corresponding test well to obtain a heat storage mode prediction model, and by predicting the heat storage mode based on the multi-scale entropy and constructing the heat storage mode prediction model, the accuracy of the heat storage mode prediction can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of geothermal development technology, and in particular to the construction of a prediction model for a thermal storage mode and a prediction method, device, medium and equipment. Background Art

[0002] The prediction of geothermal resource reserves is very important for geothermal exploitation. Sometimes, due to the uncertainty brought by the complex geological structure, there is a large deviation between the heat storage model and the heat storage prediction and the heat storage in the actual exploitation process. Single-well extraction and injection technology is an efficient geothermal exploitation and utilization technology, but the extraction and injection technology also has higher requirements for heat storage, and requires accurate prediction of the heat storage energy in the current environment. The existing heat storage prediction method based on key factors of monitoring data has a low accuracy rate due to the uncertainty of the geological environment, and it is difficult to meet the technical indicators of single-well extraction and injection. Summary of the invention

[0003] Therefore, the technical problem to be solved by the present invention is to overcome the defect of low prediction accuracy of existing thermal storage models, thereby providing a prediction model construction and prediction method, device, medium and equipment for the thermal storage model.

[0004] According to a first aspect, an embodiment of the present invention discloses a method for constructing a prediction model of a heat storage mode, the method comprising: obtaining a pressure difference signal sample set, the pressure difference signal sample set comprising pressure difference signal samples corresponding to the pump bottom of a pumping pump in a plurality of heat storage mode test wells during the pumping process; performing feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy; calculating the entropy increase rate of the multi-scale entropy, the entropy increase rate characterizing the growth rate of sample entropy values ​​at multiple time scales in the multi-scale entropy; using a training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, as well as the heat storage mode of the corresponding test well, to train a preset prediction model to obtain a heat storage mode prediction model.

[0005] Optionally, the pressure difference signal samples in the pressure difference signal sample set are subjected to feature extraction processing to obtain corresponding multi-scale entropy, including: coarse-graining the time series of the pressure difference signal samples to obtain a time series of 30 time scales; and feature extraction processing is performed on the time series of the 30 time scales to obtain 30-scale entropy.

[0006] Optionally, calculating the entropy increase rate of the multi-scale entropy includes: selecting sample entropy values ​​of the first 7 time scales among the 30 scale entropies; and calculating the sample entropy values ​​of the first 7 time scales to obtain the entropy increase rate.

[0007] Optionally, after the preset prediction model is trained using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal in the training sample set and the heat storage mode of the corresponding test well to obtain the heat storage mode prediction model, the method further includes: performing an accuracy test on the heat storage mode prediction model using the test sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the test sample set and the heat storage mode of the corresponding test well; when the accuracy test result does not meet the requirements, optimizing the parameters of the heat storage mode prediction model until the accuracy test result meets the requirements.

[0008] Optionally, the preset prediction model is a support vector machine model.

[0009] According to the second aspect, an embodiment of the present invention further discloses a method for predicting a heat storage pattern, the method comprising: obtaining a target pressure difference signal at the bottom of a pump of a target well during water pumping; extracting a target multi-scale entropy from the target pressure difference signal and calculating a target entropy increase rate; inputting the target multi-scale entropy and the target entropy increase rate into a heat storage pattern prediction model to obtain a heat storage pattern corresponding to the target well, the heat storage pattern prediction model being constructed using a prediction model of a heat storage pattern as described in the first aspect or the optional implementation method discussed in the first aspect.

[0010] According to the third aspect, an embodiment of the present invention further discloses a device for constructing a prediction model of a heat storage mode, the device comprising: a sample set acquisition module, used to acquire a pressure difference signal sample set, the pressure difference signal sample set comprising pressure difference signal samples corresponding to the bottom of a pump during the pumping process of a pumping pump in a plurality of heat storage mode test wells; a multi-scale entropy acquisition module, used to perform feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain the corresponding multi-scale entropy; an entropy increase rate calculation module, used to calculate the entropy increase rate of the multi-scale entropy, the entropy increase rate characterizing the growth rate of sample entropy values ​​at multiple time scales in the multi-scale entropy; a model training module, used to train a preset prediction model using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, as well as the heat storage mode of the corresponding test well, to obtain a heat storage mode prediction model.

[0011] According to the fourth aspect, an embodiment of the present invention further discloses a thermal storage mode prediction device, the device comprising: a target signal acquisition module, used to obtain a target pressure difference signal at the bottom of a pump of a target well during water pumping; a signal processing module, used to extract a target multi-scale entropy from the target pressure difference signal and calculate a target entropy increase rate; a thermal storage mode determination module, used to input the target multi-scale entropy and the target entropy increase rate into a thermal storage mode prediction model to obtain a thermal storage mode corresponding to the target well, the thermal storage mode prediction model being constructed as the thermal storage mode prediction model described in the first aspect or any optional embodiment of the first aspect.

[0012] According to the fifth aspect, an embodiment of the present invention further discloses an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method for constructing a prediction model of a thermal storage mode as described in the first aspect or any optional embodiment of the first aspect, or the steps of the method for predicting a thermal storage mode as described in the second aspect or any optional embodiment of the second aspect.

[0013] According to the sixth aspect, an embodiment of the present invention further discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing a prediction model for a heat storage pattern as described in the first aspect or any optional embodiment of the first aspect, or the steps of the method for predicting a heat storage pattern as described in the second aspect or any optional embodiment of the second aspect.

[0014] The technical solution of the present invention has the following advantages:

[0015] The method for constructing a prediction model of a heat storage mode provided by the present invention obtains a sample set of pressure difference signals, performs feature extraction processing on the pressure difference signal samples in the sample set to obtain multi-scale entropy, and calculates the entropy increase rate based on the multi-scale entropy. The preset prediction model is trained using a training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, as well as the heat storage mode of the corresponding test well, to obtain a heat storage mode prediction model. By predicting the heat storage mode based on the multi-scale entropy and constructing the heat storage mode prediction model, the accuracy of the heat storage mode prediction can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A flowchart of a specific example of a method for constructing a prediction model of a thermal storage mode in an embodiment of the present invention;

[0018] Figure 2 A diagram of a pressure difference sensing device of a specific example of a method for constructing a prediction model of a thermal storage mode in an embodiment of the present invention;

[0019] Figure 3 A multi-scale entropy curve diagram of a specific example of a method for constructing a prediction model of a thermal storage mode in an embodiment of the present invention;

[0020] Figure 4 is a flowchart of a specific example of a method for predicting a thermal storage mode in an embodiment of the present invention;

[0021] Figure 5 A principle block diagram of a specific example of a device for building a prediction model for a thermal storage mode in an embodiment of the present invention;

[0022] Figure 6 is a principle block diagram of a specific example of a prediction device for a thermal storage mode in an embodiment of the present invention;

[0023] Figure 7 FIG. 4 is a specific example diagram of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0026] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0027] The embodiment of the present invention discloses a method for constructing a prediction model of a heat storage mode, such as Figure 1 As shown, the method comprises the following steps:

[0028] Step 101: Acquire a pressure difference signal sample set.

[0029] Among them, the pressure difference signal sample set includes the pressure difference signal samples corresponding to the bottom of the pump during the pumping process of the deep well pumps of the test wells of various thermal storage modes;

[0030] For example, the pressure difference signal samples of various heat storage modes of the test well in the embodiment of the present application can be directly obtained from the database storing the pressure difference signals of various heat storage modes in history, or a set of pressure difference sensing devices can be installed at the bottom of the deep well pump of the test well, and the pressure difference signal samples corresponding to various heat storage modes can be obtained by changing the boundary conditions of the deep well pump, wherein, for example Figure 2 As shown, the test well includes a heat exchange unit, a well wall and an inner pipe. The differential pressure sensing device includes two measuring points, which are located at the bottom of the deep well pump and the distance between the two measuring points is 10 cm. The sensing line is placed along the well wall of the test well and connected to the differential pressure sensor at the wellhead. The differential pressure sensor can use a Monslot 3051 sensor, which can save the cost of obtaining differential pressure signal samples and is easy to operate. The differential pressure sensor response time can be set to 100ms / time, the sampling frequency to 10HZ, and the operating condition collection for 30 minutes and other parameters to collect differential pressure signal samples. For example, the heat storage mode can include a remote constant pressure replenishment mode (pumping ratio is greater than 1), which is defined as Q C (r>1), remote constant pressure supply mode (pumping ratio is less than 1), defined as Q C (r<1), remote constant flow replenishment mode, defined as H C and remote no-recharge mode, defined as Q n , just as an example.

[0031] Step 102, performing feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy;

[0032] Exemplarily, in an embodiment of the present application, after obtaining a set of pressure difference signal samples, feature extraction can be performed on the pressure difference signal samples based on multi-scale entropy, including coarse-graining the time series of the pressure difference signal samples, dividing the original time series into new time series at multiple time scales, and performing feature extraction calculations on the new time series. The multi-scale entropy corresponding to each pressure difference signal sample can be obtained. The multi-scale entropy can study the time series of the signal at different time scales, and can better reflect the nonlinear characteristics of the pressure difference signal. Therefore, predicting the thermal storage mode of the test well based on the multi-scale entropy as a characteristic parameter can improve the prediction accuracy of the prediction model of the thermal storage mode.

[0033] Step 103, calculating the entropy increase rate of the multi-scale entropy.

[0034] The entropy increase rate represents the growth rate of sample entropy values ​​at multiple time scales in the multi-scale entropy;

[0035] For example, in the embodiment of the present application, after obtaining the multi-scale entropy corresponding to each pressure difference signal sample, the sample entropy values ​​of multiple time scales in the multi-scale entropy are selected for calculation to obtain the corresponding entropy increase rate. The method for calculating the sample entropy values ​​of multiple time scales is not limited. For example, the least squares method can be used to fit and obtain the slope of the sample entropy values ​​of multiple time scales, which is the entropy increase rate, for example only.

[0036] Step 104 , training a preset prediction model using a training sample set in the pressure difference signal sample set, multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, and a thermal storage mode of the corresponding test well to obtain a thermal storage mode prediction model.

[0037] Exemplarily, in an embodiment of the present application, the pressure difference signal sample set is split according to a preset ratio to obtain a training sample set and a test sample set. For example, 70% of the samples in the pressure difference signal sample set can be used as a training sample set, and the remaining 30% of the sample set can be used as a test sample set. The pressure difference signal samples in the training sample set and the corresponding multi-scale entropy and entropy increase rate, as well as the thermal storage mode of the corresponding test well are input into a preset prediction model for training, so that the preset prediction model can classify the pressure difference signal samples into corresponding thermal storage modes according to the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, and obtain a prediction model for the thermal storage mode. The preset prediction model in an embodiment of the present application can also be a neural network model. By pre-setting parameter values ​​such as weights corresponding to the neural network, the training sample set and the multi-scale entropy and entropy increase rate corresponding to the pressure difference signal samples in the training sample set, as well as the thermal storage mode of the corresponding test well are input into the preset prediction model. By optimizing the parameter values ​​corresponding to the neural network, a prediction model for the thermal storage mode corresponding to the optimal parameter values ​​is obtained. This is only used as an example.

[0038] The method for constructing a prediction model of a heat storage mode provided by the present invention obtains a sample set of pressure difference signals, performs feature extraction processing on the pressure difference signal samples in the sample set to obtain multi-scale entropy, and calculates the entropy increase rate based on the multi-scale entropy. The preset prediction model is trained using a training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, as well as the heat storage mode of the corresponding test well, to obtain a heat storage mode prediction model. By predicting the heat storage mode based on the multi-scale entropy and constructing the heat storage mode prediction model, the accuracy of the heat storage mode prediction can be improved.

[0039] As an optional embodiment of the present invention, the pressure difference signal samples in the pressure difference signal sample set are subjected to feature extraction processing to obtain corresponding multi-scale entropy, including: coarse-graining the time series of the pressure difference signal samples to obtain a time series of 30 time scales; and feature extraction processing is performed on the time series of the 30 time scales to obtain 30-scale entropy.

[0040] For example, the embodiment of the present application performs coarse-graining processing on the time series of the pressure difference signal samples, divides the original time series into new time series of 30 time scales, performs feature extraction calculation on the time series of 30 time scales, and obtains the 30-scale entropy corresponding to each pressure difference signal sample, such as Figure 3 As shown, 30-scale entropy curves corresponding to the four heat storage modes can be obtained. Just as an example, dividing the original time series into 30 time scales can not only reflect the changing law of multi-scale entropy, but also accurately predict the corresponding heat storage mode based on multi-scale entropy, while simplifying the amount of calculation.

[0041] As a specific embodiment of the present invention, a coarse-breakpoint transformation is performed on the original one-dimensional discrete time series [μ(i), i=1, 2, ..., N], where N represents the N time points in the time series and i represents the i-th time point; L vectors x(1), x(2), ..., x(L) of length τ are obtained, where τ represents the time scale factor, L=N / τ; the mean of each vector is taken as the new time series, and a new time series of length L=N / τ can be obtained as follows:

[0042]

[0043] in, represents the mean of the j-th vector; j represents the j-th vector.

[0044] Secondly, the embedding dimension m, the value of m is generally 2 according to specific needs. Using the phase space reconstruction method, the new time series with a length of L after coarse-graining is reconstructed into an m-dimensional sub-time series. A total of L-m+1 sub-time series Y1, Y2, K, Y L-m+1 , calculate Y by the following formula τ (i) and Y τ (j) The distance with the largest difference between the corresponding elements

[0045]

[0046] Among them, d m Indicates the maximum distance of the difference; represents the mean of the i-th vector; represents the mean of the j-th vector; k represents the k-th dimension.

[0047] Given a similarity threshold r, which can generally be 0.1-0.25 times the standard deviation (SD) of the original time series, the Y i τ and The number of corresponding elements with the largest difference between them is less than r, and then the ratio of this number to the total number of distances Lm is calculated by the following formula:

[0048]

[0049] in, Represents Y i τ and The ratio of the number of corresponding elements with the largest difference less than r to the total number of distances Lm;

[0050] Secondly, we calculate the mean of all points by the following formula, denoted as B τ,m (r):

[0051]

[0052] Among them, B τ,m (r) represents the mean of all points;

[0053] Secondly, increase the dimension to m+1 and repeat the above steps to get B τ,m+1 (r).

[0054] Finally, the sample entropy value of the time series corresponding to each time scale τ after time coarsening is calculated, which is the multi-scale entropy:

[0055] MSE={τ|SpEn(τ,m,r,N)=-lnB τ,m+1 (r) / B τ,m (r)}

[0056] As an optional implementation of the present invention, calculating the entropy increase rate of the multi-scale entropy includes: selecting sample entropy values ​​of the first 7 time scales among the 30 scale entropies; calculating the sample entropy values ​​of the first 7 time scales to obtain the entropy increase rate.

[0057] For example, in the embodiment of the present application, after obtaining the 30-scale entropy corresponding to each pressure difference signal sample, the change law of the entropy increase of the first 7 scales in the 30-scale entropy is significant, which not only saves the amount of calculation, but also can accurately calculate the entropy increase rate. Therefore, the sample entropy values ​​of the first 7 time scales in the 30-scale entropy can be selected, and the least squares method can be used for fitting calculation to obtain the corresponding entropy increase rate, which is only used as an example.

[0058] As an optional embodiment of the present invention, the preset prediction model is trained using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal in the training sample set, as well as the heat storage mode of the corresponding test well. After the heat storage mode prediction model is obtained, the method further includes: performing an accuracy test on the heat storage mode prediction model using the test sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the test sample set, as well as the heat storage mode of the corresponding test well; when the accuracy test result does not meet the requirements, optimizing the parameters of the heat storage mode prediction model until the accuracy test result meets the requirements.

[0059] Exemplarily, after the embodiment of the present application obtains the heat storage mode prediction model through the above embodiment, the remaining 30% of the test sample set can be selected to test the accuracy of the heat storage mode prediction model. When the accuracy reaches the preset threshold, the heat storage mode prediction model can be used to perform mode prediction. For example, if the accuracy of the heat storage mode prediction model reaches more than 90%, it indicates that the heat storage mode prediction model meets the requirements. This is only taken as an example. If the accuracy does not reach more than 90%, the parameters of the heat storage mode prediction model in the above embodiment can be optimized until the accuracy test results meet the requirements. This is only taken as an example.

[0060] As an optional implementation manner of the present invention, the preset prediction model is a support vector machine model.

[0061] Exemplarily, the preset prediction model in the embodiment of the present application can be a support vector machine model. The support vector machine model algorithm is simple, only a small number of samples are needed to solve the classification problem, and it is robust. By pre-setting the penalty coefficient and kernel function in the vector machine model, the pressure difference signal samples and the corresponding multi-scale entropy and entropy increase rate in the training sample set and the thermal storage mode of the corresponding test well are input into the preset prediction model for training. By optimizing the penalty coefficient and kernel function in the vector machine model, the thermal storage mode prediction model is obtained. When the remaining 30% of the test sample set is selected to test the accuracy of the thermal storage mode prediction model, if the accuracy test result does not meet the requirements, the penalty coefficient and kernel function in the vector machine model will continue to be optimized until the accuracy test result meets the requirements. This is just an example.

[0062] The embodiment of the present invention also discloses a method for predicting a heat storage mode, such as Figure 4 As shown, the method includes:

[0063] Step 201, obtaining a target differential pressure signal at the bottom of a pump of a target well during a pumping process.

[0064] Illustratively, the embodiment of the present application can obtain a target differential pressure signal at the bottom of a pump of a target well during the pumping process through the differential pressure sensing device of the above embodiment.

[0065] Step 203, extracting target multi-scale entropy from the target pressure difference signal and calculating target entropy increase rate. For example, the specific method of extracting target multi-scale entropy from the target pressure difference signal and calculating target entropy increase rate is as described in the above embodiment, which will not be repeated here.

[0066] Step 204, input the target multi-scale entropy and the target entropy increase rate into the thermal storage mode prediction model to obtain the thermal storage mode corresponding to the target well, and the thermal storage mode prediction model is constructed by the thermal storage mode prediction model described in the above embodiment.

[0067] The method for predicting a heat storage mode provided by the present invention obtains a target pressure difference signal at the bottom of a pump of a target well during water pumping, extracts a target multi-scale entropy from the target pressure difference signal and calculates a target entropy increase rate, inputs the target multi-scale entropy and the target entropy increase rate into a heat storage mode prediction model, obtains a heat storage mode corresponding to the target well, inputs the target pressure difference signal to be tested into the heat storage mode prediction model, determines the corresponding heat storage mode, which can not only improve the accuracy of heat storage mode prediction, but also improve the efficiency of heat storage mode prediction.

[0068] The embodiment of the present invention also discloses a prediction model building device for a heat storage mode, such as Figure 5 As shown, the device comprises:

[0069] The sample set acquisition module 301 is used to acquire a pressure difference signal sample set, wherein the pressure difference signal sample set includes pressure difference signal samples corresponding to the bottom of the pump of the water pumping pump of the test wells of various heat storage modes during the water pumping process;

[0070] The multi-scale entropy acquisition module 302 is used to perform feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy;

[0071] An entropy increase rate calculation module 303 is used to calculate the entropy increase rate of the multi-scale entropy, where the entropy increase rate represents the growth rate of sample entropy values ​​at multiple time scales in the multi-scale entropy;

[0072] The model training module 304 is used to train the preset prediction model using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set and the heat storage mode of the corresponding test well to obtain a heat storage mode prediction model.

[0073] The device for constructing a prediction model of a heat storage mode provided by the present invention obtains a sample set of pressure difference signals, performs feature extraction processing on the pressure difference signal samples in the sample set to obtain multi-scale entropy, calculates the entropy increase rate based on the multi-scale entropy, and trains a preset prediction model using a training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set and the heat storage mode of the corresponding test well to obtain a heat storage mode prediction model. By predicting the heat storage mode based on the multi-scale entropy and constructing the heat storage mode prediction model, the accuracy of the heat storage mode prediction can be improved.

[0074] As an optional embodiment of the present invention, the multi-scale entropy acquisition module includes: a sequence processing submodule, which is used to perform coarse-graining processing on the time series of the pressure difference signal samples to obtain a time series of 30 time scales; a 30-scale entropy acquisition submodule, which is used to perform feature extraction processing on the time series of the 30 time scales to obtain 30-scale entropy.

[0075] As an optional embodiment of the present invention, the entropy increase rate calculation module includes: an entropy value selection submodule, used to select sample entropy values ​​of the first 7 time scales among the 30 scale entropies; an entropy increase rate calculation submodule, used to calculate the sample entropy values ​​of the first 7 time scales to obtain the entropy increase rate.

[0076] As an optional embodiment of the present invention, the device also includes: a model testing module, which is used to test the accuracy of the heat storage mode prediction model by using the test sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the test sample set and the heat storage mode of the corresponding test well; a parameter optimization module, which is used to optimize the parameters of the heat storage mode prediction model when the accuracy test result does not meet the requirements until the accuracy test result meets the requirements.

[0077] As an optional implementation manner of the present invention, the preset prediction model is a support vector machine model.

[0078] The embodiment of the present invention also discloses a prediction device for a heat storage mode, such as Figure 6 As shown, the device comprises:

[0079] The target signal acquisition module 501 is used to acquire the target differential pressure signal at the bottom of the pump of the target well during the pumping process;

[0080] A signal processing module 502 is used to extract target multi-scale entropy from the target pressure difference signal and calculate a target entropy increase rate;

[0081] The heat storage mode determination module 503 is used to input the target multi-scale entropy and the target entropy increase rate into the heat storage mode prediction model to obtain the heat storage mode corresponding to the target well. The heat storage mode prediction model is constructed by the heat storage mode prediction model described in the above embodiment.

[0082] The embodiment of the present invention further provides an electronic device, such as Figure 7 As shown, the electronic device may include a processor 401 and a memory 402, wherein the processor 401 and the memory 402 may be connected via a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0083] The processor 401 may be a central processing unit (CPU). The processor 401 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips. Figure 2 The processor shown can be used to change the boundary conditions of the deep well pump to obtain pressure difference signals corresponding to various heat storage modes, and can also be used to execute the heat storage mode prediction model construction method or heat storage mode prediction method of the above embodiment.

[0084] The memory 402 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the prediction model construction of the thermal storage mode or the prediction method of the thermal storage mode in the embodiment of the present invention. The processor 401 executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory 402, that is, to implement the prediction model construction of the thermal storage mode or the prediction method of the thermal storage mode in the above method embodiment.

[0085] The memory 402 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created by the processor 401, etc. In addition, the memory 402 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 402 may optionally include a memory remotely arranged relative to the processor 401, and these remote memories may be connected to the processor 401 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0086] The one or more modules are stored in the memory 402, and when executed by the processor 401, the following is performed: Figure 1 The prediction model construction method of the heat storage mode in the embodiment shown in the figure or Figure 4 The prediction method of the thermal storage mode in the illustrated embodiment.

[0087] For details of the above electronic equipment, please refer to Figure 1 or Figure 4 The corresponding related descriptions and effects in the illustrated embodiments can be understood and will not be repeated here.

[0088] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0089] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the defined scope.

Claims

1. A method for constructing a prediction model for a heat storage mode, characterized in that: The method comprises: Acquire a pressure difference signal sample set, wherein the pressure difference signal sample set includes pressure difference signal samples corresponding to the bottom of a pump of a deep well pump of a test well in multiple thermal storage modes during a pumping process; Performing feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy; Calculating the entropy increase rate of the multiscale entropy, wherein the entropy increase rate represents the growth rate of sample entropy values ​​at multiple time scales in the multiscale entropy; The preset prediction model is trained using the training sample set in the pressure difference signal sample set, the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set, and the thermal storage mode of the corresponding test well to obtain a thermal storage mode prediction model.

2. The method for constructing a prediction model of a thermal storage mode according to claim 1, characterized in that: The feature extraction process is performed on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy, including: The time series of the pressure difference signal samples are coarse-grained to obtain a time series of 30 time scales; The time series of the 30 time scales are subjected to feature extraction processing to obtain 30-scale entropy.

3. The method for constructing a prediction model of a thermal storage mode according to claim 2, characterized in that: The calculating the entropy increase rate of the multi-scale entropy comprises: Select the sample entropy values ​​of the first 7 time scales among the 30 scale entropies; The sample entropy values ​​of the first seven time scales are calculated to obtain the entropy increase rate.

4. The method according to claim 1, characterized in that: After the preset prediction model is trained using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal in the training sample set and the heat storage mode of the corresponding test well to obtain the heat storage mode prediction model, the method further includes: The accuracy of the heat storage mode prediction model is tested by using a test sample set in the pressure difference signal sample set, multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the test sample set, and a heat storage mode of a corresponding test well; When the accuracy test result does not meet the requirements, the parameters of the thermal storage mode prediction model are optimized until the accuracy test result meets the requirements.

5. The method according to claim 1, characterized in that The preset prediction model is a support vector machine model.

6. A method for predicting a thermal storage mode, characterized in that: The method comprises: Obtain a target differential pressure signal at the bottom of a deep well pump of a target well during the pumping process; Extracting target multi-scale entropy from the target pressure difference signal and calculating target entropy increase rate; The target multi-scale entropy and the target entropy increase rate are input into a thermal storage mode prediction model to obtain a thermal storage mode corresponding to the target well, and the thermal storage mode prediction model is constructed by the thermal storage mode prediction model according to any one of claims 1 to 5.

7. A device for constructing a prediction model of a heat storage mode, characterized in that: The device comprises: A sample set acquisition module, used to acquire a pressure difference signal sample set, wherein the pressure difference signal sample set includes pressure difference signal samples corresponding to the bottom of a deep well pump of a test well in a plurality of thermal storage modes during the pumping process; A multi-scale entropy acquisition module is used to perform feature extraction processing on the pressure difference signal samples in the pressure difference signal sample set to obtain corresponding multi-scale entropy; An entropy increase rate calculation module, used to calculate the entropy increase rate of the multi-scale entropy, wherein the entropy increase rate represents the growth rate of sample entropy values ​​at multiple time scales in the multi-scale entropy; The model training module is used to train the preset prediction model using the training sample set in the pressure difference signal sample set and the multi-scale entropy and entropy increase rate corresponding to each pressure difference signal sample in the training sample set and the thermal storage mode of the corresponding test well to obtain a thermal storage mode prediction model.

8. A prediction device for heat storage mode, characterized in that: The device comprises: A target signal acquisition module, used to acquire a target differential pressure signal at the bottom of a deep well pump of a target well during the pumping process; A signal processing module, used for extracting target multi-scale entropy from the target pressure difference signal and calculating a target entropy increase rate; A heat storage mode determination module is used to input the target multi-scale entropy and the target entropy increase rate into a heat storage mode prediction model to obtain the heat storage mode corresponding to the target well, and the heat storage mode prediction model is constructed by the heat storage mode prediction model described in any one of claims 1 to 5.

9. An electronic device, characterized in that: include: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the steps of the method for constructing a prediction model of a thermal storage mode as described in any one of claims 1 to 5 or the method for predicting a thermal storage mode as described in claim 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for constructing a prediction model of a thermal storage mode as described in any one of claims 1 to 5 or the method for predicting a thermal storage mode as described in claim 6 are implemented.