Air source heat pump load prediction method and system based on big data

Abstract

The invention relates to the field of load prediction, in particular to an air source heat pump load prediction method and system based on big data. Indoor and outdoor temperature data are subjected to time sequence analysis by adopting an artificial intelligence technology based on deep learning, time sequence change characteristics of indoor temperature and outdoor temperature are respectively captured, indoor and outdoor temperature difference characteristics are calculated according to the time sequence change characteristics, and a time sequence change mode of indoor and outdoor temperature difference in a global time domain is mined; therefore, the electric load prediction of the air source heat pump user is carried out, and the influence of the real-time environmental factors on the electric load of the air source heat pump can be more accurately captured, so that the load prediction precision is improved, and more reliable data support is provided for optimal scheduling and energy management of a power system.

Description

Technical Field

[0001] The present application relates to the field of load forecasting, and specifically to a method and system for air source heat pump load forecasting based on big data. Background Art

[0002] As an efficient and environmentally friendly energy conversion device, the main function of air source heat pump is to provide winter heating services for buildings. Its core principle is to absorb low-grade heat energy in the air and increase it to high-grade heat energy to meet the indoor temperature requirements. Therefore, the power load of air source heat pump is closely related to the user's demand for heating, forming a dynamic interactive relationship.

[0003] Specifically, when the outdoor temperature drops, the heat energy in the air decreases, and the air source heat pump needs to consume more electricity to extract and increase this heat energy to keep the indoor temperature stable, that is, its electrical load increases. On the contrary, if the outdoor temperature rises, the heat energy is abundant, the working efficiency of the heat pump is improved, and the electrical load will decrease accordingly. This phenomenon reflects the inverse proportional relationship between the electrical load of the air source heat pump and the outdoor temperature.

[0004] In practical applications, in order to accurately predict and control the operating status of the power system and optimize energy scheduling and management, it is particularly important to accurately predict the load of air source heat pumps. However, traditional air source heat pump load prediction methods mainly rely on empirical formulas and statistical models, and only roughly consider seasonal factors, resulting in low prediction accuracy and difficulty in meeting the needs of modern energy management systems for high-precision and real-time predictions.

[0005] Therefore, in order to improve the accuracy of air source heat pump load prediction, an optimized air source heat pump load prediction method is needed. Summary of the invention

[0006] This application is made in consideration of the above problems. One purpose of this application is to provide a method and system for air source heat pump load prediction based on big data.

[0007] The embodiment of the present application provides an air source heat pump load prediction method based on big data, which includes:

[0008] Obtaining a time queue of outdoor temperature and a time queue of indoor temperature collected by a temperature sensor;

[0009] Performing time series feature extraction on the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors;

[0010] Based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors, extracting indoor and outdoor temperature difference features to obtain a sequence of indoor and outdoor temperature difference time series feature vectors;

[0011] Performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series;

[0012] Based on the indoor and outdoor temperature difference time series significant aggregation representation vector, an air source heat pump user electric load prediction decoding value is generated.

[0013] For example, according to the air source heat pump load prediction method based on big data of an embodiment of the present application, the time series feature extraction is performed on the time series of the outdoor temperature and the time series of the indoor temperature to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors, including:

[0014] The time queue of the outdoor temperature and the time queue of the indoor temperature are input into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of the outdoor temperature time series associated feature vectors and a sequence of the indoor temperature time series associated feature vectors.

[0015] For example, according to the air source heat pump load prediction method based on big data of an embodiment of the present application, based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors, the indoor and outdoor temperature difference features are extracted to obtain a sequence of indoor and outdoor temperature difference time series feature vectors, including:

[0016] The differential feature vectors between each corresponding group of outdoor temperature time series association feature vectors and indoor temperature time series association feature vectors in the sequence of the outdoor temperature time series association feature vectors and the sequence of the indoor temperature time series association feature vectors are calculated to obtain the sequence of the indoor and outdoor temperature differential time series feature vectors.

[0017] For example, according to the air source heat pump load prediction method based on big data in an embodiment of the present application, the sequence of the indoor and outdoor temperature difference time series feature vectors is forward propagated and aggregated to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series, including:

[0018] The sequence of indoor and outdoor temperature difference time series feature vectors is input into a feature forward propagation aggregation module based on node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0019] For example, according to the air source heat pump load prediction method based on big data of an embodiment of the present application, the sequence of the indoor and outdoor temperature difference time series feature vectors is input into a feature forward propagation aggregation module based on the node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector, including:

[0020] Based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of indoor and outdoor temperature difference time series feature vectors, calculating the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector to obtain a time series of energy significant descriptors;

[0021] Extracting the timestamp of each indoor and outdoor temperature difference time series feature vector in the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a time series of timestamps;

[0022] The indoor and outdoor temperature difference time series feature vector corresponding to the current time point in the sequence of the indoor and outdoor temperature difference time series feature vectors is used as the current indoor and outdoor temperature difference time series feature vector, and the indoor and outdoor temperature difference time series feature vectors corresponding to other time points are used as the historical indoor and outdoor temperature difference time series feature vectors, so as to obtain a sequence of the current indoor and outdoor temperature difference time series feature vector and the historical indoor and outdoor temperature difference time series feature vector;

[0023] Calculating the energy time decay factor of each historical indoor and outdoor temperature difference time series feature vector in the sequence of the historical indoor and outdoor temperature difference time series feature vectors relative to the current indoor and outdoor temperature difference time series feature vector to obtain a time series of energy time decay factors;

[0024] Using the time series of the energy significant descriptor as a positive adjustment factor and the time series of the energy time decay factor as a negative adjustment factor, the sequence of the historical indoor and outdoor temperature difference time series feature vectors is subjected to time series propagation aggregation to obtain a historical node indoor and outdoor temperature difference time series aggregation feature vector;

[0025] The indoor and outdoor temperature difference time series aggregation feature vector of the historical node and the current indoor and outdoor temperature difference time series feature vector are fused to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0026] For example, according to the air source heat pump load prediction method based on big data of an embodiment of the present application, based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of the indoor and outdoor temperature difference time series feature vector, the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector is calculated to obtain a time series of energy significant descriptors, including:

[0027] The expected value of the fourth power of the difference between each eigenvalue in the indoor and outdoor temperature difference time series eigenvalue vector and its eigenmean is calculated, and the expected value is divided by the square of the eigenvariance of the indoor and outdoor temperature difference time series eigenvalue vector to obtain the energy significant descriptor.

[0028] For example, according to the air source heat pump load prediction method based on big data in an embodiment of the present application, the energy time decay factor of each historical indoor and outdoor temperature difference time series feature vector in the sequence of the historical indoor and outdoor temperature difference time series feature vectors relative to the current indoor and outdoor temperature difference time series feature vector is calculated to obtain a time series of energy time decay factors, including:

[0029] Calculate the timestamp difference between the historical indoor and outdoor temperature difference time series feature vector and the current indoor and outdoor temperature difference time series feature vector;

[0030] The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient;

[0031] An exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent is calculated to obtain the energy time decay factor.

[0032] For example, according to the air source heat pump load prediction method based on big data in an embodiment of the present application, the air source heat pump user power load prediction decoding value is generated based on the indoor and outdoor temperature difference time series significant aggregation representation vector, including:

[0033] The indoor and outdoor temperature difference time series significant aggregation representation vector is input into the decoder-based air source heat pump user electric load prediction module to obtain the air source heat pump user electric load prediction decoding value.

[0034] The embodiment of the present application also provides an air source heat pump load prediction system based on big data, which includes:

[0035] A temperature acquisition module, used to acquire a time queue of outdoor temperature and a time queue of indoor temperature acquired by a temperature sensor;

[0036] A time series feature extraction module, used to extract time series features from the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of time series associated feature vectors of the outdoor temperature and a sequence of time series associated feature vectors of the indoor temperature;

[0037] An indoor and outdoor temperature difference feature extraction module, used to extract indoor and outdoor temperature difference features based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors to obtain a sequence of indoor and outdoor temperature difference time series feature vectors;

[0038] An aggregation module, used for performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series;

[0039] The prediction module is used to generate a predicted decoding value of the electric load of the air source heat pump user based on the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0040] For example, according to the air source heat pump load prediction system based on big data in an embodiment of the present application, the time series feature extraction module is used to:

[0041] The time queue of the outdoor temperature and the time queue of the indoor temperature are input into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of the outdoor temperature time series associated feature vectors and a sequence of the indoor temperature time series associated feature vectors.

[0042] The present application has at least the following technical effects: Compared with the prior art, the air source heat pump load prediction method and system based on big data of the embodiment of the present application adopts artificial intelligence technology based on deep learning to perform time series analysis on indoor and outdoor temperature data, respectively captures the time series change characteristics of indoor and outdoor temperatures, and calculates the temperature difference characteristics indoors and outdoors, and mines the time series change pattern of indoor and outdoor temperature differences in the global time domain, so as to predict the electric load of air source heat pump users, and can more accurately capture the impact of real-time environmental factors on the electric load of air source heat pumps, thereby improving the accuracy of load prediction and providing more reliable data support for the optimal scheduling and energy management of power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0044] Figure 1 A schematic diagram of the application architecture of the air source heat pump load prediction method based on big data in an embodiment of the present application is shown;

[0045] Figure 2 A flow chart of a method for predicting air source heat pump load based on big data in an embodiment of the present application is shown;

[0046] Figure 3 A flowchart of sub-step S540 of the air source heat pump load prediction method based on big data in an embodiment of the present application is shown;

[0047] Figure 4A schematic diagram of the structure of an air source heat pump load prediction system based on big data in an embodiment of the present application is shown; and

[0048] Figure 5 An application scenario diagram of the air source heat pump load prediction method based on big data in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0050] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0051] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0052] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0053] Figure 1 A schematic diagram of the application architecture of an air source heat pump load prediction method based on big data in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0054] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0055] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0056] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the program of the air source heat pump load prediction method based on big data in the embodiment of the present application.

[0057] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the air source heat pump load prediction methods based on big data in the embodiments of the present application according to the obtained program instructions.

[0058] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0059] The following is a non-restrictive explanation of the air source heat pump load prediction method based on big data provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0060] In response to the above technical problems, the technical concept of the present application is to use artificial intelligence technology based on deep learning to perform time series analysis on indoor and outdoor temperature data, capture the time series change characteristics of indoor and outdoor temperatures respectively, and use this to calculate the temperature difference characteristics between indoor and outdoor, and explore the time series change pattern of indoor and outdoor temperature differences in the global time domain. In this way, the electric load of air source heat pump users can be predicted, which can more accurately capture the impact of real-time environmental factors on the electric load of air source heat pumps, thereby improving the accuracy of load prediction and providing more reliable data support for the optimal scheduling and energy management of power systems.

[0061] Based on this, Figure 2 The flowchart of the air source heat pump load prediction method based on big data in the embodiment of the present application is shown. For example, the air source heat pump load prediction method based on big data can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, according to the air source heat pump load prediction method based on big data according to the embodiment of the present application, the steps include: S510, obtaining the time queue of outdoor temperature and the time queue of indoor temperature collected by the temperature sensor; S520, respectively performing time series feature extraction on the time queue of the outdoor temperature and the time queue of the indoor temperature to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors; S530, based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors, extracting indoor and outdoor temperature difference features to obtain a sequence of indoor and outdoor temperature difference time series feature vectors; S540, performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain an indoor and outdoor temperature difference time series significant aggregation representation vector; S550, generating an air source heat pump user power load prediction decoding value based on the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0062] Specifically, in the technical solution of the present application, first, a time queue of outdoor temperature and a time queue of indoor temperature collected by a temperature sensor are obtained. In an embodiment of the present application, outdoor temperature data and indoor temperature data are collected in hours as a time unit to obtain a time queue of outdoor temperature and a time queue of indoor temperature.

[0063] Considering that the indoor and outdoor temperatures are data that changes dynamically over time, in order to fully understand the changing trends of the outdoor and indoor temperatures, this application uses a temperature sequence encoder based on a Bi-LSTM model to perform temporal encoding on the time queue of the outdoor temperature and the time queue of the indoor temperature to extract their temporal change characteristics. Among them, the Bi-LSTM model uses a bidirectional LSTM unit to capture the forward and backward temporal context dependencies of the temperature data of each time step to generate the temporal correlation features of each time step, and obtain a sequence of outdoor temperature temporal correlation feature vectors and a sequence of indoor temperature temporal correlation feature vectors, so as to describe the evolution of outdoor temperature and indoor temperature over time.

[0064] Correspondingly, in step S520, time series feature extraction is performed on the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors, including: inputting the time queue of the outdoor temperature and the time queue of the indoor temperature into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors.

[0065] Next, since the load of the air source heat pump (i.e., power consumption or heating / cooling capacity) is closely related to the temperature difference between indoors and outdoors, in order to more intuitively reveal the temperature difference between indoors and outdoors, thereby providing a more effective data basis for subsequent load forecasting, in the technical solution of the present application, the differential feature vectors between each group of corresponding outdoor temperature time series correlation feature vectors and indoor temperature time series correlation feature vectors in the sequence of the outdoor temperature time series correlation feature vectors and the sequence of the indoor temperature time series correlation feature vectors are further calculated to obtain a sequence of indoor and outdoor temperature difference time series feature vectors, thereby characterizing the dynamic changes of the indoor and outdoor temperature difference over time, reflecting the changing trend and amplitude of the temperature difference, and providing a more intuitive and effective feature description for the prediction model.

[0066] Correspondingly, in step S530, based on the sequence of the outdoor temperature time series associated feature vectors and the sequence of the indoor temperature time series associated feature vectors, the indoor and outdoor temperature difference characteristics are extracted to obtain a sequence of indoor and outdoor temperature difference time series feature vectors, including: calculating the differential feature vector between each group of corresponding outdoor temperature time series associated feature vectors and indoor temperature time series associated feature vectors in the sequence of the outdoor temperature time series associated feature vectors and the sequence of the indoor temperature time series associated feature vectors to obtain the sequence of the indoor and outdoor temperature difference time series feature vectors.

[0067] Secondly, in order to capture the temporal variation pattern of the indoor and outdoor temperature difference characteristics in the global time domain, it is necessary to further perform feature aggregation on the sequence of the indoor and outdoor temperature difference time series feature vectors. Based on this, in the technical solution of the present application, a feature forward propagation aggregation module based on the node energy time series significance is introduced to perform adaptive feature aggregation processing on the sequence of the indoor and outdoor temperature difference time series feature vectors. Specifically, the module first calculates the energy significant descriptor according to the statistical characteristic parameters of each indoor and outdoor temperature difference time series feature vector, quantitatively describes its information density and feature importance, and assigns time information to the indoor and outdoor temperature difference time series feature vector of each time node, and uses the time series decay mechanism to generate the corresponding energy time decay factor based on the time span between the indoor and outdoor temperature difference time series feature vector of each historical time node and the current (i.e., the end of the time series) indoor and outdoor temperature difference time series feature vector, so that in the dynamic aggregation process of feature forward propagation, information can be aggregated based on the feature significance and feature time series decay characteristics of each time node to generate a significant aggregation representation vector of the indoor and outdoor temperature difference time series.

[0068] Correspondingly, in step S540, the sequence of indoor and outdoor temperature difference time series feature vectors is subjected to forward time propagation aggregation to obtain an indoor and outdoor temperature difference time series significant aggregation representation vector, including: inputting the sequence of indoor and outdoor temperature difference time series feature vectors into a feature forward propagation aggregation module based on node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0069] Specifically, Figure 3 As shown, the sequence of the indoor and outdoor temperature difference time series feature vectors is input into the feature forward propagation aggregation module based on the node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector, including: S541, based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of the indoor and outdoor temperature difference time series feature vectors, the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector is calculated to obtain the time series of energy significant descriptors; S542, the timestamps of each indoor and outdoor temperature difference time series feature vector in the sequence of the indoor and outdoor temperature difference time series feature vector are extracted to obtain the time series of timestamps; S543, the indoor and outdoor temperature difference time series feature vector corresponding to the current time point in the sequence of the indoor and outdoor temperature difference time series feature vector is used as the current indoor and outdoor temperature difference time series feature vector, and the indoor and outdoor temperature difference time series feature vectors corresponding to other time points are used as the historical indoor and outdoor temperature difference time series feature vectors. outdoor temperature difference time series feature vector to obtain the current indoor and outdoor temperature difference time series feature vector and the sequence of historical indoor and outdoor temperature difference time series feature vectors; S544, calculate the energy time decay factor of each historical indoor and outdoor temperature difference time series feature vector in the sequence of historical indoor and outdoor temperature difference time series feature vectors relative to the current indoor and outdoor temperature difference time series feature vector to obtain the time series of energy time decay factors; S545, using the time series of the energy significant descriptor as the positive adjustment factor and the time series of the energy time decay factor as the negative adjustment factor, perform time series propagation aggregation on the sequence of historical indoor and outdoor temperature difference time series feature vectors to obtain the historical node indoor and outdoor temperature difference time series aggregation feature vector; S546, fuse the historical node indoor and outdoor temperature difference time series aggregation feature vector and the current indoor and outdoor temperature difference time series feature vector to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0070] Among them, in step S541, based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of the indoor and outdoor temperature difference time series feature vector, the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector is calculated to obtain a time series of energy significant descriptors, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the indoor and outdoor temperature difference time series feature vector and its eigenvalue mean, and dividing the expected value by the square of the eigenvariance of the indoor and outdoor temperature difference time series feature vector to obtain the energy significant descriptor.

[0071] Among them, in step S544, the energy time decay factor of each historical indoor and outdoor temperature difference time series characteristic vector in the sequence of the historical indoor and outdoor temperature difference time series characteristic vectors relative to the current indoor and outdoor temperature difference time series characteristic vector is calculated to obtain a time series of the energy time decay factor, including: calculating the timestamp difference between the historical indoor and outdoor temperature difference time series characteristic vector and the current indoor and outdoor temperature difference time series characteristic vector; rounding the timestamp difference down and dividing it by a preset time decay period and then squaring it, and multiplying the square result by a preset decay rate to obtain an initial energy decay coefficient; calculating an exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent to obtain the energy time decay factor.

[0072] Inputting the sequence of indoor and outdoor temperature difference time series feature vectors into a feature forward propagation aggregation module based on node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector, including: processing the sequence of indoor and outdoor temperature difference time series feature vectors with the following time series propagation fusion formula to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector; wherein the time series propagation fusion formula is:

[0073] X={x1,x2,...x k-1 ,x k}

[0074]

[0075] Wherein, X represents the sequence of the indoor and outdoor temperature difference time series feature vectors, the value of k is the number of feature vectors in the sequence of the indoor and outdoor temperature difference time series feature vectors, x1, x2, x i 、x k-1 and x k respectively represent the first, second, ith, k-1th and current indoor and outdoor temperature difference time series feature vectors in the sequence of the indoor and outdoor temperature difference time series feature vectors, represents the jth eigenvalue of the i-th indoor and outdoor temperature difference time series eigenvector, the value of L is the length of the i-th indoor and outdoor temperature difference time series eigenvector, μi is the characteristic mean of the ith indoor and outdoor temperature difference time series characteristic vector, σ 2 i is the characteristic variance of the ith indoor and outdoor temperature difference time series characteristic vector, represents the fourth-order central moment of the ith indoor and outdoor temperature difference time series eigenvector, that is, the expected value of the fourth power of the difference between each eigenvalue in the ith indoor and outdoor temperature difference time series eigenvector and its eigenmean, is the energy significant descriptor of the i-th indoor and outdoor temperature difference time series feature vector, is the energy significant descriptor of the current indoor and outdoor temperature difference time series feature vector, t k The timestamp representing the current indoor and outdoor temperature difference time series feature vector, t i represents the timestamp of the i-th indoor and outdoor temperature difference time series feature vector, represents rounding down, β represents the preset decay rate, C represents the preset time decay period, exp(·) represents the exponential function operation, α is the weight hyperparameter, and x k+1 The indoor and outdoor temperature difference time series significant aggregation representation vector is represented.

[0076] Furthermore, the indoor and outdoor temperature difference time series significant aggregation representation vector is input into the decoder-based air source heat pump user electric load prediction module to obtain the air source heat pump user electric load prediction decoded value. That is, the decoder model is used to feature decode and reconstruct the time evolution information of the temperature difference in the historical time steps contained in the indoor and outdoor temperature difference time series significant aggregation representation vector to predict the electric load of the air source heat pump user. In the technical solution of the present application, the decoder has undergone a large amount of training optimization, continuously adjusted the model parameters, and effectively learned the complex nonlinear relationship between the indoor and outdoor temperature difference and the electric load, so that the indoor and outdoor temperature difference time series significant aggregation representation vector can be accurately decoded and predicted to output the electric load prediction decoded value.

[0077] Correspondingly, based on the indoor and outdoor temperature difference time series significant aggregation representation vector, an air source heat pump user electric load prediction decoding value is generated, including: inputting the indoor and outdoor temperature difference time series significant aggregation representation vector into a decoder-based air source heat pump user electric load prediction module to obtain the air source heat pump user electric load prediction decoding value.

[0078] Preferably, the indoor and outdoor temperature difference time series significant aggregation representation vector is input into a decoder-based air source heat pump user electricity load prediction module to obtain the air source heat pump user electricity load prediction decoding value, including: subtracting the 0 norm of the indoor and outdoor temperature difference time series significant aggregation representation vector from the length of the indoor and outdoor temperature difference time series significant aggregation representation vector to obtain the indoor and outdoor temperature difference time series significant aggregation isolated representation value; calculating a power function with each eigenvalue of the indoor and outdoor temperature difference time series significant aggregation representation vector as the base, and the difference of the indoor and outdoor temperature difference time series significant aggregation isolated representation value minus one as the exponent, and multiplying it by the base 2 logarithm of the product of the indoor and outdoor temperature difference time series significant aggregation isolated representation value plus one and the indoor and outdoor temperature difference time series significant aggregation isolated representation value to obtain the result. Obtain a leading value of indoor and outdoor temperature difference time series significant aggregation; multiply each eigenvalue of the indoor and outdoor temperature difference time series significant aggregation representation vector by the difference of the indoor and outdoor temperature difference time series significant aggregation isolated representation value minus one and divide it by the indoor and outdoor temperature difference time series significant aggregation isolated representation value, and then calculate an exponential function with a natural constant as the base to obtain an indoor and outdoor temperature difference time series significant aggregation bias value; add the indoor and outdoor temperature difference time series significant aggregation leading value to the indoor and outdoor temperature difference time series significant aggregation bias value to obtain each eigenvalue of the optimized indoor and outdoor temperature difference time series significant aggregation representation vector; input the optimized indoor and outdoor temperature difference time series significant aggregation representation vector into a decoder-based air source heat pump user electric load prediction module to obtain an air source heat pump user electric load prediction decoding value.

[0079] Specifically, the indoor and outdoor temperature difference time series significant aggregation representation vector, for example, each eigenvalue v of V is recorded as i , where v i The optimization process of ∈V is expressed as follows:

[0080]

[0081] n=l-‖V‖ C

[0082] Wherein, V represents the significant aggregation representation vector of the indoor and outdoor temperature difference time series, v i represents the ith eigenvalue of the indoor and outdoor temperature difference time series significant aggregation representation vector, l represents the length of the indoor and outdoor temperature difference time series significant aggregation representation vector, ‖·‖0 represents the zero norm of the vector, n represents the indoor and outdoor temperature difference time series significant aggregation isolated representation value, log[·] represents the logarithm operation with base 2, exp(·) represents the exponential operation with base e, (·) ⊙(n-1) It represents the power function of calculating the eigenvalues ​​of each position in the eigenvector with the difference value minus one as the exponent of the significant aggregation isolated representation value of the indoor and outdoor temperature difference time series, v i ′ represents the i-th eigenvalue of the indoor and outdoor temperature difference time series significant aggregation representation vector after the optimization.

[0083] Among them, the sequence of indoor and outdoor temperature difference time series feature vectors respectively expresses the local time series correlation characteristics of the indoor and outdoor temperature difference. Therefore, when performing forward propagation aggregation of the node energy time series significance characteristics, the full time domain time series offset caused by the reinforcement difference under each time series node is taken into account. It is expected to further improve the regression comprehensibility of the full time series aggregation sequence of the indoor and outdoor temperature difference time series significance aggregation representation vector, thereby improving the accuracy of the air source heat pump user electric load prediction decoding value obtained by inputting the decoder-based air source heat pump user electric load prediction module.

[0084] Therefore, for the high-dimensional feature manifold of the indoor and outdoor temperature difference time series significant aggregation representation vector, the eigenvalues ​​of the feature set are used as the vector field representation of the aggregation dimension, and the value of the derivative of the vector field of the indoor and outdoor temperature difference time series significant aggregation representation vector at the isolated zero position is used as the order information to fix the local position of the eigenvalues ​​of its feature set, and a bias for the reversibility of the feature regression distribution of the indoor and outdoor temperature difference time series significant aggregation representation vector is added as a reward to achieve the mapping target tracking of the regression distribution of the indoor and outdoor temperature difference time series significant aggregation representation vector for the eigenvalue position, so that the feature set of the indoor and outdoor temperature difference time series significant aggregation representation vector can perceive the mapping migration to the aggregation distribution, thereby improving the accuracy of the air source heat pump user electric load prediction decoding value obtained by inputting the indoor and outdoor temperature difference time series significant aggregation representation vector into the decoder-based air source heat pump user electric load prediction module.

[0085] Based on the above embodiments, see Figure 4 As shown, it is a structural schematic diagram of an air source heat pump load prediction system 800 based on big data in an embodiment of the present application. The air source heat pump load prediction system 800 based on big data includes: a temperature acquisition module 810, which is used to obtain the time queue of outdoor temperature and the time queue of indoor temperature collected by the temperature sensor; a time series feature extraction module 820, which is used to extract the time series features of the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors; an indoor and outdoor temperature difference feature extraction module 830, which is used to extract the indoor and outdoor temperature difference features based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors to obtain a sequence of indoor and outdoor temperature difference time series feature vectors; an aggregation module 840, which is used to perform forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain an indoor and outdoor temperature difference time series significant aggregation representation vector; a prediction module 850, which is used to generate an air source heat pump user power load prediction decoding value based on the indoor and outdoor temperature difference time series significant aggregation representation vector.

[0086] In one example, in the above-mentioned big data-based air source heat pump load prediction system 800, the time series feature extraction module 820 is used to: input the time queue of the outdoor temperature and the time queue of the indoor temperature into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of the outdoor temperature time series associated feature vectors and a sequence of the indoor temperature time series associated feature vectors.

[0087] The specific functions and operations of each module in the air source heat pump load prediction system 800 based on big data have been referred to above. Figure 2 to Figure 3 The description of the air source heat pump load prediction method based on big data has been introduced in detail, and therefore, its repeated description will be omitted.

[0088] Figure 5 FIG. 1 is an application scenario diagram of the air source heat pump load prediction method based on big data according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, the time queue of the outdoor temperature collected by the temperature sensor is obtained (for example, Figure 5 D1) and the time sequence of indoor temperature (e.g., Figure 5 Then, the time queue of the outdoor temperature and the time queue of the indoor temperature are input to a server (for example, Figure 5 In S) shown in , the server is able to use the air source heat pump load prediction algorithm based on big data to process the time queue of the outdoor temperature and the time queue of the indoor temperature to obtain the air source heat pump user power load prediction decoding value.

[0089] Based on the above embodiments, another exemplary embodiment of an electronic device is provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the air source heat pump load prediction method based on big data in the above embodiments when executing the program. Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0090] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the air source heat pump load prediction method based on big data according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0091] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the air source heat pump load prediction method based on big data according to the embodiment of the present application.

[0092] Those skilled in the art will appreciate that various variations and improvements may occur in the content disclosed in the present application. For example, various devices or components described above may be implemented by hardware, or by some or all of the combination of software, firmware, or the three. In addition, although the application has made various references to some units in the system according to an embodiment of the application, any number of different units may be used and run on a client and / or server. The unit is only illustrative, and the different aspects of the system and method may use different units.

[0093] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0094] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0095] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. The air source heat pump load prediction method based on big data is characterized by: include: Obtaining a time queue of outdoor temperature and a time queue of indoor temperature collected by a temperature sensor; Performing time series feature extraction on the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors; Based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors, extracting indoor and outdoor temperature difference features to obtain a sequence of indoor and outdoor temperature difference time series feature vectors; Performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series; Based on the indoor and outdoor temperature difference time series significant aggregation representation vector, an air source heat pump user electric load prediction decoding value is generated.

2. The air source heat pump load prediction method based on big data according to claim 1 is characterized in that: The time series feature extraction is performed on the time series of the outdoor temperature and the time series of the indoor temperature respectively to obtain a sequence of outdoor temperature time series associated feature vectors and a sequence of indoor temperature time series associated feature vectors, including: The time queue of the outdoor temperature and the time queue of the indoor temperature are input into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of the outdoor temperature time series associated feature vectors and a sequence of the indoor temperature time series associated feature vectors.

3. The air source heat pump load prediction method based on big data according to claim 2 is characterized in that: Based on the sequence of the outdoor temperature time series associated feature vectors and the sequence of the indoor temperature time series associated feature vectors, extracting indoor and outdoor temperature difference features to obtain a sequence of indoor and outdoor temperature difference time series feature vectors, including: The differential feature vectors between each corresponding group of outdoor temperature time series association feature vectors and indoor temperature time series association feature vectors in the sequence of the outdoor temperature time series association feature vectors and the sequence of the indoor temperature time series association feature vectors are calculated to obtain the sequence of the indoor and outdoor temperature differential time series feature vectors.

4. The air source heat pump load prediction method based on big data according to claim 3 is characterized in that: Performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series, including: The sequence of indoor and outdoor temperature difference time series feature vectors is input into a feature forward propagation aggregation module based on node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

5. The air source heat pump load prediction method based on big data according to claim 4 is characterized in that: Inputting the sequence of indoor and outdoor temperature difference time series feature vectors into a feature forward propagation aggregation module based on node energy time series significance to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector, including: Based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of indoor and outdoor temperature difference time series feature vectors, calculating the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector to obtain a time series of energy significant descriptors; Extracting the timestamp of each indoor and outdoor temperature difference time series feature vector in the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a time series of timestamps; The indoor and outdoor temperature difference time series feature vector corresponding to the current time point in the sequence of the indoor and outdoor temperature difference time series feature vectors is used as the current indoor and outdoor temperature difference time series feature vector, and the indoor and outdoor temperature difference time series feature vectors corresponding to other time points are used as the historical indoor and outdoor temperature difference time series feature vectors, so as to obtain a sequence of the current indoor and outdoor temperature difference time series feature vector and the historical indoor and outdoor temperature difference time series feature vector; Calculating the energy time decay factor of each historical indoor and outdoor temperature difference time series feature vector in the sequence of the historical indoor and outdoor temperature difference time series feature vectors relative to the current indoor and outdoor temperature difference time series feature vector to obtain a time series of energy time decay factors; Using the time series of the energy significant descriptor as a positive adjustment factor and the time series of the energy time decay factor as a negative adjustment factor, the sequence of the historical indoor and outdoor temperature difference time series feature vectors is subjected to time series propagation aggregation to obtain a historical node indoor and outdoor temperature difference time series aggregation feature vector; The indoor and outdoor temperature difference time series aggregation feature vector of the historical node and the current indoor and outdoor temperature difference time series feature vector are fused to obtain the indoor and outdoor temperature difference time series significant aggregation representation vector.

6. The air source heat pump load prediction method based on big data according to claim 5 is characterized in that: Based on the mean and variance of each indoor and outdoor temperature difference time series feature vector in the sequence of the indoor and outdoor temperature difference time series feature vectors, calculating the energy significant descriptor of each indoor and outdoor temperature difference time series feature vector to obtain a time series of energy significant descriptors, including: The expected value of the fourth power of the difference between each eigenvalue in the indoor and outdoor temperature difference time series eigenvalue vector and its eigenmean is calculated, and the expected value is divided by the square of the eigenvariance of the indoor and outdoor temperature difference time series eigenvalue vector to obtain the energy significant descriptor.

7. The air source heat pump load prediction method based on big data according to claim 6 is characterized in that: Calculating the energy time decay factor of each historical indoor and outdoor temperature difference time series feature vector in the sequence of the historical indoor and outdoor temperature difference time series feature vectors relative to the current indoor and outdoor temperature difference time series feature vector to obtain a time series of energy time decay factors, including: Calculate the timestamp difference between the historical indoor and outdoor temperature difference time series feature vector and the current indoor and outdoor temperature difference time series feature vector; The timestamp difference is rounded down and divided by a preset time decay period and then squared, and the square result is multiplied by a preset decay rate to obtain an initial energy decay coefficient; An exponential function value with a natural constant as the base and the initial energy decay coefficient as the exponent is calculated to obtain the energy time decay factor.

8. The air source heat pump load prediction method based on big data according to claim 7 is characterized in that: Based on the indoor and outdoor temperature difference time series significant aggregation representation vector, the air source heat pump user electric load prediction decoding value is generated, including: The indoor and outdoor temperature difference time series significant aggregation representation vector is input into the decoder-based air source heat pump user electric load prediction module to obtain the air source heat pump user electric load prediction decoding value.

9. The air source heat pump load prediction system based on big data is characterized by: include: A temperature acquisition module, used to acquire a time queue of outdoor temperature and a time queue of indoor temperature acquired by a temperature sensor; A time series feature extraction module, used to extract time series features from the time queue of the outdoor temperature and the time queue of the indoor temperature respectively to obtain a sequence of time series associated feature vectors of the outdoor temperature and a sequence of time series associated feature vectors of the indoor temperature; An indoor and outdoor temperature difference feature extraction module, used to extract indoor and outdoor temperature difference features based on the sequence of outdoor temperature time series associated feature vectors and the sequence of indoor temperature time series associated feature vectors to obtain a sequence of indoor and outdoor temperature difference time series feature vectors; An aggregation module, used for performing forward time series propagation aggregation on the sequence of indoor and outdoor temperature difference time series feature vectors to obtain a significant aggregation representation vector of the indoor and outdoor temperature difference time series; The prediction module is used to generate a predicted decoding value of the electric load of the air source heat pump user based on the indoor and outdoor temperature difference time series significant aggregation representation vector.

10. The air source heat pump load prediction system based on big data according to claim 9, characterized in that: The temporal feature extraction module is used to: The time queue of the outdoor temperature and the time queue of the indoor temperature are input into a temperature sequence encoder based on a Bi-LSTM model to obtain a sequence of the outdoor temperature time series associated feature vectors and a sequence of the indoor temperature time series associated feature vectors.

Images