A power supply method and system for a water heater

Abstract

A power supply method and system for a water heater are disclosed. The system includes: a water heater body; a solar panel installed on or near the top of the water heater body; an inverter connected between the solar panel and the water heater body for converting direct current (DC) to alternating current (AC) and adjusting the output voltage and frequency according to the operating state of the water heater body; an energy storage device connected between the inverter and the water heater body for storing excess AC power and providing supplementary power to the water heater body when solar energy is insufficient; a controller connected between the inverter and the energy storage device for monitoring and controlling the operating parameters of the power supply system; and a communication line connected between the controller and the water heater body for transmitting operating data and instructions of the power supply system to achieve coordination between the power supply system and the water heater body. This improves energy utilization efficiency.

Description

Technical Field

[0001] The present application relates to the field of water heaters, and more specifically, to a power supply method and system for a water heater. Background Art

[0002] A domestic water heater is a common household appliance used to heat water supply. Traditional water heaters are usually powered by the power grid, but with the development of renewable energy, solar panels are gradually being used in the power supply system of water heaters. Solar panels can convert solar energy into direct current, but water heaters usually require alternating current to work properly. Therefore, a power supply system is needed to convert the direct current generated by solar panels into alternating current suitable for the operation of water heaters.

[0003] An inverter is an electronic device connected between a solar panel and a water heater to convert direct current into alternating current. However, traditional inverters usually have fixed output characteristics and cannot be flexibly adjusted according to the needs of the load, that is, traditional inverters are usually designed to operate at a specific voltage and frequency and cannot be dynamically adjusted according to changes in the load. This may cause the output voltage and frequency to not match the load requirements, affecting the performance and effect of the water heater. In addition, there are some existing inverters that can control the output voltage and frequency, but due to their limited control accuracy, their output voltage and frequency may have a certain error. This means that even if the inverter can be adjusted to a certain extent, it still cannot fully meet the precise needs of the load, and the error range may cause the water heater to work unstable or have unsatisfactory results.

[0004] Therefore, an optimized power supply system for a water heater is desired. Summary of the invention

[0005] In view of this, the present application proposes a power supply method and system for a water heater, which can automatically adjust the output voltage and frequency according to the working state of the water heater, thereby improving the efficiency and stability of the water heater.

[0006] According to one aspect of the present application, a power supply system for a water heater is provided, comprising:

[0007] Hot water machine body;

[0008] A solar panel mounted on or near the top of the water heater body for converting sunlight into direct current electricity;

[0009] An inverter connected between the solar panel and the water heater body, used to convert direct current into alternating current and adjust the output voltage and frequency according to the working state of the water heater body;

[0010] An energy storage device connected between the inverter and the water heater body, for storing excess AC power and providing supplementary power to the water heater body when solar energy is insufficient;

[0011] A controller connected between the inverter and the energy storage device, used to monitor and control operating parameters of the power supply system; and

[0012] A communication line connected between the controller and the water heater body is used to transmit operating data and instructions of the power supply system to achieve coordination between the power supply system and the water heater body.

[0013] According to another aspect of the present application, a power supply method for a water heater is provided. The power supply method for a water heater is applied to the power supply system for the water heater as described above, and comprises:

[0014] Obtaining heating power values ​​of the water heater main body at multiple predetermined time points within a predetermined time period;

[0015] Arranging the heating power values ​​at the plurality of predetermined time points into a heating power time series input vector according to the time dimension;

[0016] Performing local time series characteristic analysis on the heating power time series input vector to obtain a sequence of heating power local time series characteristic vectors;

[0017] Performing correlation analysis on any two adjacent local time series feature vectors of heating power in the sequence of local time series feature vectors of heating power to obtain a local time series fluctuation feature vector of heating power;

[0018] splicing the sequence of the local time series characteristic vectors of the heating power to obtain the full time series characteristic vector of the heating power;

[0019] Fusion of the heating power local time series fluctuation feature vector and the heating power full time series feature vector to obtain a multi-dimensional and multi-scale time series feature of the heating power; and

[0020] Based on the multi-dimensional and multi-scale time series characteristics of the heating power, the recommended output voltage value and frequency value are determined.

[0021] According to an embodiment of the present application, the system includes: a water heater body; a solar panel installed on the top or near the water heater body; an inverter connected between the solar panel and the water heater body, used to convert direct current into alternating current, and adjust the output voltage and frequency according to the working state of the water heater body; an energy storage device connected between the inverter and the water heater body, used to store excess alternating current and provide supplementary power to the water heater body when solar energy is insufficient; a controller connected between the inverter and the energy storage device, used to monitor and control the operating parameters of the power supply system; a communication line connected between the controller and the water heater body, used to transmit the operating data and instructions of the power supply system, so as to achieve coordination between the power supply system and the water heater body. In this way, energy utilization efficiency can be improved.

[0022] Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the present application and, together with the description, serve to explain the principles of the present application.

[0024] Figure 1 A block diagram of the inverter in the power supply system of the water heater according to an embodiment of the present application is shown.

[0025] Figure 2 A flow chart showing a method for supplying power to a water heater according to an embodiment of the present application is shown.

[0026] Figure 3 A schematic diagram of the architecture of a method for powering a water heater according to an embodiment of the present application is shown.

[0027] Figure 4 An application scenario diagram of a power supply system for a water heater according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0028] 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.

[0029] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0030] Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0031] In addition, in order to better illustrate the present application, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present application can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present application.

[0032] In view of the above technical problems, in the technical solution of the present application, a power supply system for a water heater is proposed, which includes a water heater body, a solar panel, an inverter, an energy storage device, a controller and a communication line, wherein the inverter is connected between the solar panel and the water heater, and is used to convert direct current into alternating current, and automatically adjust the output voltage and frequency according to the working state of the water heater to improve the efficiency and stability of the water heater.

[0033] In the process of the inverter automatically adjusting the output voltage and frequency according to the working state of the water heater, the technical concept of the present application is to collect the heating power value of the water heater body in real time, and introduce data processing and analysis algorithms at the back end to perform time series analysis of the heating power of the water heater, so as to recommend appropriate output voltage and frequency values. In this way, the output voltage and frequency can be automatically adjusted according to the working state of the water heater, thereby realizing sustainable power supply of the water heater and improving energy utilization efficiency.

[0034] Specifically, the power supply system of the water heater includes: a water heater body; a solar panel installed on the top or near the water heater body, used to convert sunlight into direct current; an inverter connected between the solar panel and the water heater body, used to convert direct current into alternating current and adjust the output voltage and frequency according to the working state of the water heater body; an energy storage device connected between the inverter and the water heater body, used to store excess alternating current and provide supplementary power to the water heater body when solar energy is insufficient; a controller connected between the inverter and the energy storage device, used to monitor and control the operating parameters of the power supply system; and a communication line connected between the controller and the water heater body, used to transmit operating data and instructions of the power supply system to achieve coordination between the power supply system and the water heater body.

[0035] Figure 1 FIG. 1 is a block diagram of a power supply system of a water heater according to an embodiment of the present application. Figure 1 As shown, according to the power supply system of the water heater of the embodiment of the present application, the inverter 100 includes: a data acquisition module 110, which is used to obtain the heating power value of the water heater body at multiple predetermined time points within a predetermined time period; a heating power time series arrangement module 120, which is used to arrange the heating power values ​​of the multiple predetermined time points into a heating power time series input vector according to the time dimension; a heating power local time series feature extraction module 130, which is used to perform local time series feature analysis on the heating power time series input vector to obtain a sequence of heating power local time series feature vectors; a heating power local time series fluctuation module 140, which is used to analyze the sequence of the heating power local time series feature vectors A correlation analysis is performed on any two adjacent local time series feature vectors of heating power to obtain a local time series fluctuation feature vector of heating power; a heating power full time series feature splicing module 150 is used to splice the sequence of the local time series feature vectors of heating power to obtain a full time series feature vector of heating power; a heating power multi-dimensional and multi-scale time series feature fusion module 160 is used to fuse the local time series fluctuation feature vector of heating power and the full time series feature vector of heating power to obtain a multi-dimensional and multi-scale time series feature of heating power; and an output parameter recommendation module 170 is used to determine the recommended output voltage value and frequency value based on the multi-dimensional and multi-scale time series features of heating power.

[0036] It should be understood that the data acquisition module 110 is responsible for collecting the heating power data of the water heater body; the heating power time series arrangement module 120 arranges the collected heating power values ​​in chronological order to form a heating power time series input vector; the heating power local time series feature extraction module 130 analyzes the heating power time series input vector, extracts the local time series features, and forms a sequence of heating power local time series feature vectors; the heating power local time series fluctuation module 140 is used to analyze the correlation between two adjacent feature vectors in the sequence of the heating power local time series feature vectors to obtain the heating power local time series feature vector sequence. The power local time series fluctuation feature vector; the heating power full time series feature splicing module 150 is used to splice the sequence of the heating power local time series feature vectors into a heating power full time series feature vector; the heating power multi-dimensional multi-scale time series feature fusion module 160 is used to fuse the heating power local time series fluctuation feature vector representing the local time series feature and the heating power full time series feature vector representing the full time series feature to obtain the heating power multi-dimensional multi-scale time series feature; the output parameter recommendation module 170 uses the multi-dimensional multi-scale time series feature to determine the recommended values ​​of the inverter output voltage and frequency. These modules together constitute the inverter 100, which realizes the functions of collecting the heating power of the water heater body, extracting features, and recommending parameters.

[0037] Specifically, in the technical solution of the present application, first, the heating power values ​​of the water heater body at multiple predetermined time points within a predetermined time period are obtained. Next, considering that the heating power value of the water heater body has a time-series dynamic change law in the time dimension, that is, there is a time-series correlation relationship between the heating power values ​​at the multiple predetermined time points. Therefore, in order to be able to perform time-series analysis and characterize the change trend of the heating power of the water heater, it is necessary to further arrange the heating power values ​​at the multiple predetermined time points according to the time dimension as a heating power time-series input vector, so as to integrate the time-series distribution information of the heating power value.

[0038] Furthermore, in order to better understand and analyze the changing trends and characteristics of the heating power, in the technical solution of the present application, the heating power time series input vector is vector segmented and then passed through a time series feature extractor to obtain a sequence of local time series feature vectors of the heating power. It should be understood that by vector segmenting the heating power time series input vector, it can be divided into multiple smaller sub-vectors, each sub-vector representing the heating power data over a period of time. In this way, the timing changes of the heating power can be observed in a more fine-grained manner to better understand the working mode and requirements of the load. Moreover, by processing through the time series feature extractor to obtain a sequence of local time series feature vectors of the heating power, a more comprehensive understanding of the dynamic changes of the heating power can be achieved, and more information can be provided for subsequent analysis and decision-making, thereby achieving more accurate output voltage and frequency regulation.

[0039] Correspondingly, the heating power local time series feature extraction module 130 is used to: after vector segmentation of the heating power time series input vector, obtain a sequence of the heating power local time series feature vectors through a time series feature extractor. It should be understood that the role of vector segmentation in this context is to divide the heating power time series input vector into smaller subsequences, and the purpose of doing so is to extract local time series features, that is, to extract its feature information for each subsequence separately. The benefit of vector segmentation is that it can capture local patterns and dynamic changes in the heating power time series data. By segmenting the input vector into multiple subsequences, feature extraction can be performed on each subsequence, so as to better capture the changes in heating power in different time periods. For example, if the heating power time series data represents a heating process, there may be different stages or specific events in the heating process. By segmenting the input vector into multiple subsequences, feature extraction can be performed on each subsequence separately to obtain local time series features of each stage or event. This helps to better understand the characteristics and changing laws of different stages or events in the heating process. Therefore, vector segmentation can improve the feature extraction capability of heating power time series data, so that the model can better understand and analyze the local time series characteristics in the heating process.

[0040] Subsequently, the sequence of the local time series feature vectors of the heating power is spliced ​​to obtain the full time series feature vector of the heating power, wherein the full time series feature vector of the heating power contains the complete time series change information of the heating power in the entire heating process, which is used to more deeply understand the characteristics of the heating process, and can more comprehensively describe the change trend and characteristics of the heating power, and provide more references for subsequent adjustments and optimizations.

[0041] Then, considering that the heating power of the water heater has volatility in the time dimension, this volatility plays an important role in the time series change trend of the heating power and the working state and demand of the load. Therefore, in order to better describe the volatility and change trend of the heating power, in the technical solution of the present application, the correlation between any two adjacent local time series feature vectors of the heating power in the sequence of the local time series feature vectors of the heating power is further calculated to obtain the local time series fluctuation feature vector of the heating power. It should be understood that by calculating the correlation between two adjacent local time series feature vectors of the heating power, the similarity or correlation between them can be measured to capture the volatility and change trend of the heating power, such as the amplitude, frequency, periodicity, etc. of the fluctuation, as well as the correlation between adjacent time series feature vectors. In this way, dynamic change information about the heating power can be provided to help analyze the working state and demand of the load, and adjust the output voltage and frequency of the inverter to meet the needs of the load.

[0042] Correspondingly, the heating power local time series fluctuation module 140 is used to calculate the correlation between any two adjacent heating power local time series feature vectors in the sequence of the heating power local time series feature vectors using the following correlation formula to obtain multiple correlations; wherein the correlation formula is:

[0043]

[0044] Wherein, V1 and V2 respectively represent any two adjacent local time series feature vectors of heating power in the sequence of the local time series feature vectors of heating power, and are two different linear transformations, (·) T represents the transpose operation, Representing each correlation among the multiple correlations; and arranging the multiple correlations to obtain the local time series fluctuation characteristic vector of the heating power.

[0045] Next, it is considered that the local time series fluctuation characteristic vector of the heating power describes the volatility and change trend of the heating power within the local time window. It can provide information about the instantaneous fluctuation of the heating power, such as the amplitude, frequency, periodicity, etc. of the fluctuation. These characteristics are helpful in analyzing the instantaneous changes and instability in the heating process. The full time series characteristic vector of the heating power contains the complete time series change information of the heating power in the entire heating process. It can provide the global time series characteristics of the heating power, which can reflect the overall change trend and stability of the heating power. Therefore, in order to provide richer and more comprehensive heating power feature information, in the technical solution of the present application, the local time series fluctuation characteristic vector of the heating power and the full time series characteristic vector of the heating power are further integrated to obtain a multi-dimensional and multi-scale time series characteristic vector of the heating power. In this way, the multi-dimensional and multi-scale time series characteristic vector of the heating power can simultaneously describe the instantaneous volatility and overall time series characteristics of the heating power, thereby providing a more comprehensive, rich and accurate description of the heating power characteristics.

[0046] Correspondingly, the heating power multi-dimensional and multi-scale time series feature fusion module 160 is used to: fuse the heating power local time series fluctuation feature vector and the heating power full time series feature vector to obtain the heating power multi-dimensional and multi-scale time series feature vector as the heating power multi-dimensional and multi-scale time series feature.

[0047] Then, the multi-dimensional and multi-scale time series feature vector of the heating power is passed through a decoder to obtain a decoded value, which is used to represent the recommended output voltage and frequency values. In other words, the multi-dimensional and multi-scale time series feature information fused between the instantaneous fluctuation characteristics of the heating power and the overall time series change characteristics is used to perform decoding regression to recommend appropriate output voltage and frequency values. In this way, the output voltage and frequency can be automatically adjusted according to the working state of the water heater, thereby achieving sustainable power supply of the water heater and improving energy utilization efficiency.

[0048] Correspondingly, the output parameter recommendation module 170 is used to: pass the heating power multi-dimensional multi-scale time series feature vector through a decoder to obtain a decoded value, and the decoded value is used to represent the recommended output voltage value and frequency value.

[0049] In the technical solution of the present application, the local time series fluctuation feature vector of the heating power expresses the adjacent local time domain correlation of the time series correlation features of the heating power value in the local time domain determined by vector segmentation in the global time domain, and the full time series feature vector of the heating power expresses the adjacent local time domain connectivity of the time series correlation features of the heating power value in the local time domain. Therefore, when the local time series fluctuation feature vector of the heating power and the full time series feature vector of the heating power are fused, if the sharing of the time series correlation features of the local time series fluctuation feature vector of the heating power and the full time series feature vector of the heating power can be improved, the time series correlation expression effect of the multi-dimensional and multi-scale time series feature vector of the heating power can be improved.

[0050] That is, considering the differences in the distribution of time series correlation features corresponding to different time series dimensions and scales, it is necessary to avoid the sparseness of the feature sharing distribution of local overflow features during the fusion of the time series correlation feature sharing between the local time series fluctuation feature vector of the heating power and the full time series feature vector of the heating power.

[0051] Based on this, in the technical solution of the present application, in a preferred embodiment, the multi-dimensional and multi-scale time series feature vector of the heating power is passed through a decoder to obtain a decoded value, including: calculating the mean matrix and variance matrix of the multi-dimensional and multi-scale time series feature vector of the heating power, wherein the value of each position of the mean matrix is ​​the mean of a pair of eigenvalues ​​of two positions of the multi-dimensional and multi-scale time series feature vector of the heating power corresponding to the position coordinates respectively, and the value of each position of the variance matrix is ​​the variance of a pair of eigenvalues ​​of two positions of the multi-dimensional and multi-scale time series feature vector of the heating power corresponding to the position coordinates respectively; matrix multiplying the transposed vector of the multi-dimensional and multi-scale time series feature vector of the heating power by the mean matrix to obtain a first intermediate vector and converting the obtained The variance matrix is ​​matrix-multiplied with the heating power multi-dimensional multi-scale time series feature vector to obtain a second intermediate vector, wherein the heating power multi-dimensional multi-scale time series feature vector is a column vector; the point sum of the transposed vectors of the first intermediate vector and the second intermediate vector is calculated to obtain a third intermediate vector; the matrix product of the mean matrix and the variance matrix is ​​calculated, and the transposed vector of the heating power multi-dimensional multi-scale time series feature vector and the matrix product are matrix-multiplied to obtain a fourth intermediate vector; the point sum of the transposed vectors of the third intermediate vector and the fourth intermediate vector is calculated to obtain an optimized heating power multi-dimensional multi-scale time series feature vector; the optimized heating power multi-dimensional multi-scale time series feature vector is passed through a decoder to obtain a decoded value.

[0052] That is, by using the mean matrix and variance matrix of the group aggregation statistical evaluation of the local distribution of the eigenvalue granularity of the multi-dimensional multi-scale time series feature vector of the heating power as the retrieval and response distribution enhancement of the multi-dimensional multi-scale time series feature vector of the heating power, and constructing a reference-free distribution retrieval response framework in an open domain based on the group aggregation feature distribution of the multi-dimensional multi-scale time series feature vector of the heating power, the distribution response redundancy caused by the local overflow characteristics of the multi-dimensional multi-scale time series feature vector of the heating power can be avoided through response superposition, so as to realize the fidelity constraint of the response self-aggregation statistics related to the multi-dimensional multi-scale time series feature vector of the heating power as a fusion feature representation in the decoding regression domain, thereby improving the accuracy of the decoded value obtained by the decoder of the multi-dimensional multi-scale time series feature vector of the heating power.

[0053] It is worth mentioning that the decoder is a model component used to convert the input feature vector into an understandable and usable output value. In the output parameter recommendation module, the decoder is used to convert the multi-dimensional and multi-scale time series feature vector of the heating power into a recommended output voltage value and frequency value. The decoder can learn the relationship between the feature and the output based on the input feature vector, and map the feature vector back to the original output space through the inverse operation. It can convert the abstract feature representation into a specific output result. In the output parameter recommendation module, the decoder's task is to generate a decoded value, that is, a recommended output voltage value and frequency value, based on the multi-dimensional and multi-scale time series feature vector of the heating power. By learning the mapping relationship between the input feature vector and the output, the decoder can predict the most appropriate output parameter based on the input feature vector. The decoder can convert the features learned by the model into practical and usable output suggestions. Through the decoder, the model can generate recommended output parameters based on the input feature vector to help users make decisions or perform corresponding operations. The design and training of the decoder is a key link in the recommendation system, which directly affects the output quality and accuracy of the model.

[0054] Furthermore, the power supply system of the water heater further includes a training module for training the time series feature extractor and the decoder. It should be understood that the role of the training module in the power supply system of the water heater is to train the time series feature extractor and the decoder so that they can effectively learn and infer the characteristics and output parameters of the heating power. Specifically, the training module is used to train the time series feature extractor. During the training process, by providing known heating power time series data and corresponding labels (such as expected output voltage values ​​and frequency values), the training module can adjust the parameters of the time series feature extractor so that it can extract useful features from the original time series data. Through repeated training and optimization, the time series feature extractor can gradually learn better feature representations to better capture the time series characteristics of the heating power. The training module is also used to train the decoder. The training process of the decoder is similar to that of the time series feature extractor, but its goal is to learn the mapping relationship between the feature vector and the output parameter. By providing known feature vectors and corresponding output parameter labels, the training module can adjust the parameters of the decoder so that it can accurately predict the recommended output voltage value and frequency value. Through repeated training and optimization, the decoder can gradually improve the accuracy and reliability of the output parameters. The existence of the training module enables the time series feature extractor and decoder to learn and optimize through a large amount of training data to gradually improve their performance and effect. Through the training module, the power supply system of the water heater can be adaptively trained according to actual data, thereby providing more accurate and reliable heating power feature extraction and output parameter recommendation functions.

[0055] In one example, the training module includes: a training data acquisition unit, which is used to acquire training data, wherein the training data includes the training heating power values ​​of the water heater body at multiple predetermined time points within a predetermined time period, and the real values ​​of the recommended output voltage value and frequency value; a training heating power timing arrangement unit, which is used to arrange the training heating power values ​​of the multiple predetermined time points into a training heating power timing input vector according to the time dimension; a training heating power local timing feature extraction unit, which is used to perform local timing feature analysis on the training heating power timing input vector to obtain a sequence of training heating power local timing feature vectors; a training heating power local timing fluctuation unit, which is used to perform correlation analysis on any two adjacent training heating power local timing feature vectors in the sequence of training heating power local timing feature vectors to obtain a training heating power local timing fluctuation feature vector; a training heating power full timing feature extraction unit, which is used to perform correlation analysis on any two adjacent training heating power local timing feature vectors in the sequence of training heating power local timing feature vectors to obtain a training heating power local timing fluctuation feature vector; A feature splicing unit, used to splice the sequence of the training heating power local time series feature vectors to obtain the training heating power full time series feature vector; a training heating power multi-dimensional and multi-scale time series feature fusion unit, used to fuse the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector to obtain the training heating power multi-dimensional and multi-scale time series feature vector; a decoding loss function value calculation unit, used to pass the training heating power multi-dimensional and multi-scale time series feature vector through the decoder to obtain a decoding loss function value; a specific loss function value calculation unit, used to calculate the specific loss function value of the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector; and a loss training unit, used to train the time series feature extractor and the decoder using the weighted sum of the decoding loss function value and the specific loss function value as the loss function value.

[0056] Furthermore, the decoding loss function value calculation unit is used to: use the decoder to perform decoding regression on the training heating power multi-dimensional and multi-scale time series feature vector using the following decoding training formula to obtain a training decoding value; wherein the decoding training formula is: Wherein, X is the multi-dimensional and multi-scale time series feature vector of the training heating power, Y is the training decoding value, W is the weight matrix, represents matrix multiplication; and, calculating the variance between the training decoded value and the true value as the decoding loss function value.

[0057] In particular, in the technical solution of the present application, taking into account the differences in the distribution of time series correlation features corresponding to different time series dimensions and scales, it is necessary to avoid the sparseness of the feature sharing distribution of the time series correlation features shared between the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector during fusion. Based on this, the applicant of the present application has designed a specific loss function for the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector.

[0058] Accordingly, in one example, the specific loss function value calculation unit is further used to: calculate the specific loss function value of the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector by a specific loss formula; wherein the specific loss formula is:

[0059]

[0060] Wherein, V1 is the local time series fluctuation feature vector of the training heating power, and V2 is the full time series feature vector of the training heating power, ||·||1 and ||·||2 are the 1-norm and 2-norm of the feature vector respectively, ε is the threshold hyperparameter, and the feature vectors are all in the form of row vectors, represents vector multiplication, represents vector subtraction, (·) T represents the transpose operation, is the specific loss function value.

[0061] Specifically, the strengthening of the shared key features between the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector can be regarded as the distribution information compression of the global feature set. By reconstructing the relative shape relationship of the original feature manifold based on the structural representation between the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector, the distribution sparseness control of the key features can be performed. While strengthening the shared key features between the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector, the training heating power multi-dimensional multi-scale time series feature vector can be obtained as a sparse but meaningful fusion manifold geometric representation of the fusion feature, so as to improve the time-related fusion expression effect of the training heating power multi-dimensional multi-scale time series feature vector on the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector, thereby improving the accuracy of the decoded value obtained by the training heating power multi-dimensional multi-scale time series feature vector through the decoder. In this way, the output voltage and frequency can be automatically adjusted according to the working state of the water heater, thereby realizing the sustainable power supply of the water heater and improving the energy utilization efficiency.

[0062] In summary, the power supply system of the water heater according to the embodiment of the present application is explained, which can improve energy utilization efficiency.

[0063] As described above, the power supply system of the water heater according to the embodiment of the present application can be implemented in various terminal devices, such as a server with a power supply algorithm for the water heater. In one example, the power supply system of the water heater can be integrated into the terminal device as a software module and / or a hardware module. For example, the power supply system of the water heater can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the power supply system of the water heater can also be one of the many hardware modules of the terminal device.

[0064] Alternatively, in another example, the power supply system of the water heater and the terminal device may be separate devices, and the power supply system of the water heater may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0065] Furthermore, the present application also provides a power supply method for a water heater, and the power supply method for a water heater is applied to a power supply system for a water heater as described above. Figure 2 A flow chart showing a method for supplying power to a water heater according to an embodiment of the present application is shown. Figure 3 A schematic diagram showing the system architecture of the power supply method for a water heater according to an embodiment of the present application is shown. Figure 2 and Figure 3 As shown, according to the power supply method of the water heater of the embodiment of the present application, it includes: S110, obtaining the heating power value of the water heater body at multiple predetermined time points within a predetermined time period; S120, arranging the heating power values ​​of the multiple predetermined time points into a heating power timing input vector according to the time dimension; S130, performing local timing feature analysis on the heating power timing input vector to obtain a sequence of heating power local timing feature vectors; S140, performing correlation analysis on any two adjacent heating power local timing feature vectors in the sequence of the heating power local timing feature vectors to obtain a heating power local timing fluctuation feature vector; S150, splicing the sequence of the heating power local timing feature vectors to obtain a heating power full timing feature vector; S160, fusing the heating power local timing fluctuation feature vector and the heating power full timing feature vector to obtain a heating power multi-dimensional and multi-scale timing feature; and, S170, determining the recommended output voltage value and frequency value based on the heating power multi-dimensional and multi-scale timing feature.

[0066] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned water heater power supply method have been described in the above reference. Figure 1The power supply system of the water heater has been described in detail, and therefore, its repeated description will be omitted.

[0067] Figure 4 FIG. 1 shows an application scenario diagram of a power supply system for a water heater according to an embodiment of the present application. Figure 4 As shown, in this application scenario, first, the heating power values ​​of the water heater body at multiple predetermined time points within a predetermined time period are obtained (for example, Figure 4 D), then, the heating power values ​​at the plurality of predetermined time points are input to a server (eg, Figure 4 In S) shown in , the server is able to use the power supply algorithm of the water heater to process the heating power values ​​of the multiple predetermined time points to obtain decoded values ​​for representing recommended output voltage values ​​and frequency values.

[0068] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the function involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a dedicated hardware-based system that performs the function or action of the specification, or can be realized by a combination of special-purpose hardware and computer instructions.

[0069] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A power supply system for a water heater, characterized in that: include: Hot water machine body; A solar panel mounted on or near the top of the water heater body for converting sunlight into direct current electricity; An inverter connected between the solar panel and the water heater body, used to convert direct current into alternating current and adjust the output voltage and frequency according to the working state of the water heater body; An energy storage device connected between the inverter and the water heater body, for storing excess AC power and providing supplementary power to the water heater body when solar energy is insufficient; A controller connected between the inverter and the energy storage device, used to monitor and control operating parameters of the power supply system; as well as A communication line connected between the controller and the water heater body, for transmitting operation data and instructions of the power supply system to achieve coordination between the power supply system and the water heater body; Wherein, the inverter comprises: A data acquisition module, used to obtain heating power values ​​of the water heater body at multiple predetermined time points within a predetermined time period; A heating power time series arrangement module, used for arranging the heating power values ​​at the plurality of predetermined time points into a heating power time series input vector according to the time dimension; A heating power local time series feature extraction module, used for performing local time series feature analysis on the heating power time series input vector to obtain a sequence of heating power local time series feature vectors; A heating power local time series fluctuation module is used to perform correlation analysis on any two adjacent heating power local time series feature vectors in the sequence of the heating power local time series feature vectors to obtain a heating power local time series fluctuation feature vector; A heating power full time series feature splicing module, used for splicing the sequence of the heating power local time series feature vectors to obtain a heating power full time series feature vector; A heating power multi-dimensional and multi-scale time series feature fusion module, used for fusing the heating power local time series fluctuation feature vector and the heating power full time series feature vector to obtain the heating power multi-dimensional and multi-scale time series feature; and The output parameter recommendation module is used to determine the recommended output voltage value and frequency value based on the multi-dimensional and multi-scale time series characteristics of the heating power.

2. The power supply system of the water heater according to claim 1, characterized in that: The heating power local time series feature extraction module is used to: The heating power time series input vector is vector segmented and then passed through a time series feature extractor to obtain a sequence of the heating power local time series feature vectors.

3. The power supply system of the water heater according to claim 2, characterized in that: The heating power local time series fluctuation module is used for: The correlation between any two adjacent heating power local time series feature vectors in the sequence of the heating power local time series feature vectors is calculated using the following correlation formula to obtain multiple correlations; wherein the correlation formula is: in, and respectively represent any two adjacent local time series feature vectors of heating power in the sequence of the local time series feature vectors of heating power, and are two different linear transformations, represents the transpose operation, indicating each of the plurality of correlations; and The multiple correlations are arranged to obtain the local time series fluctuation characteristic vector of the heating power.

4. The power supply system of the water heater according to claim 3, characterized in that: The heating power multi-dimensional and multi-scale time series feature fusion module is used to: The heating power local time series fluctuation feature vector and the heating power full time series feature vector are fused to obtain a heating power multi-dimensional and multi-scale time series feature vector as the heating power multi-dimensional and multi-scale time series feature.

5. The power supply system of the water heater according to claim 4, characterized in that: The output parameter recommendation module is used to: The multi-dimensional and multi-scale time series feature vector of the heating power is passed through a decoder to obtain a decoded value, and the decoded value is used to represent the recommended output voltage value and frequency value.

6. The power supply system of the water heater according to claim 5, characterized in that: It also includes a training module for training the temporal feature extractor and the decoder.

7. The power supply system of the water heater according to claim 6, characterized in that: The training module comprises: A training data acquisition unit, used to acquire training data, wherein the training data includes training heating power values ​​of the water heater body at multiple predetermined time points within a predetermined time period, and real values ​​of the recommended output voltage value and frequency value; A training heating power time series arrangement unit, used for arranging the training heating power values ​​at the plurality of predetermined time points into a training heating power time series input vector according to a time dimension; A training heating power local time series feature extraction unit, used for performing local time series feature analysis on the training heating power time series input vector to obtain a sequence of training heating power local time series feature vectors; A training heating power local time series fluctuation unit is used to perform a correlation analysis on any two adjacent training heating power local time series feature vectors in the sequence of the training heating power local time series feature vectors to obtain a training heating power local time series fluctuation feature vector; A training heating power full time series feature splicing unit, used for splicing the sequence of the training heating power local time series feature vectors to obtain a training heating power full time series feature vector; A training heating power multi-dimensional and multi-scale time series feature fusion unit, used for fusing the training heating power local time series fluctuation feature vector and the training heating power full time series feature vector to obtain a training heating power multi-dimensional and multi-scale time series feature vector; A decoding loss function value calculation unit, used for passing the training heating power multi-dimensional multi-scale time series feature vector through the decoder to obtain a decoding loss function value; a specific loss function value calculation unit, used to calculate the specific loss function value of the local time series fluctuation feature vector of the training heating power and the full time series feature vector of the training heating power; and A loss training unit is used to train the temporal feature extractor and the decoder by using the weighted sum of the decoding loss function value and the specific loss function value as the loss function value.

8. The power supply system of the water heater according to claim 7, characterized in that: The decoding loss function value calculation unit is used to: The decoder is used to perform decoding regression on the training heating power multi-dimensional multi-scale time series feature vector using the following decoding training formula to obtain a training decoding value; wherein the decoding training formula is: ;in, is the multi-dimensional and multi-scale time series feature vector of the training heating power, is the training decoded value, is the weight matrix, represents matrix multiplication; and The variance between the training decoded value and the true value is calculated as the decoding loss function value.

9. A method for powering a water heater, the method for powering a water heater being applied to a power supply system for a water heater as claimed in any one of claims 1 to 8, characterized in that: include: Obtaining heating power values ​​of the water heater main body at multiple predetermined time points within a predetermined time period; Arranging the heating power values ​​at the plurality of predetermined time points into a heating power time series input vector according to the time dimension; Performing local time series characteristic analysis on the heating power time series input vector to obtain a sequence of heating power local time series characteristic vectors; Performing correlation analysis on any two adjacent local time series feature vectors of heating power in the sequence of local time series feature vectors of heating power to obtain a local time series fluctuation feature vector of heating power; splicing the sequence of the local time series characteristic vectors of the heating power to obtain the full time series characteristic vector of the heating power; The heating power local time series fluctuation feature vector and the heating power full time series feature vector are integrated to obtain a multi-dimensional and multi-scale time series feature of the heating power; as well as Based on the multi-dimensional and multi-scale time series characteristics of the heating power, the recommended output voltage value and frequency value are determined.

Images