Comprehensive energy data fusion method, device and equipment

Abstract

The present invention relates to the field of integrated energy technology, and its implementation method provides an integrated energy data fusion method, device and equipment. The integrated energy data fusion method comprises: obtaining integrated energy data; the integrated energy data comprises metering data, operating status data or monitoring data generated by energy sensing equipment; using a trained echo state network to determine the business type to which each integrated energy data belongs; extracting the data content of each integrated energy data according to the business type of each integrated energy data; and fusing the integrated energy data according to the semantic association feature quantity corresponding to the data content. The implementation method provided by the present invention can improve the fusion efficiency of integrated energy data.

Description

Technical Field

[0001] The present invention relates to the field of integrated energy technology, and in particular to an integrated energy data fusion method, an integrated energy data fusion device and an integrated energy data fusion equipment. Background Art

[0002] At the perception layer, power grid equipment obtains basic information about the power grid status through multiple types of sensors and acquisition terminals, and uses information and communication technologies such as the Internet of Things to build data links to achieve data interconnection, intercommunication, and sharing, and to carry out multi-dimensional integration and comprehensive analysis of data. With the advancement of the construction of urban integrated energy systems, there are more and more intelligent business types such as distributed energy access control, distribution automation, "multi-meter integration" information collection, and orderly control of electric vehicle charging and discharging. The number of equipment is increasing, the communication technology available at the local communication access layer is complex, and the communication requirements of various businesses have not yet been unified. The underlying connection protocols of the equipment communication network vary greatly, and the communication flexibility and reliability are limited. There are certain restrictions, which restrict the realization of interconnection and intercommunication; the data of each device forms an information island and cannot form a full data network. The storage and processing of massive related data requires reliable business perception, calculation, analysis, and processing models to better support the real-time analysis and intelligent processing of integrated energy local business.

[0003] Traditional data collection and status perception terminals do not have data analysis capabilities. Data is connected to unified data through information collection terminals at all levels to achieve centralized calculation and analysis. It cannot meet the high real-time requirements of business such as real-time fault analysis and regional energy autonomy. In addition, a large number of existing sensor devices have not been fully utilized, user energy consumption information is not perceived in a timely manner, distribution information coverage is incomplete, and emerging business perception is not fully shared. Each professional system is self-contained and sensor equipment is deployed repeatedly, making it difficult to achieve one-time collection and sharing. The master data models of equipment in various business scenarios are not unified, and data between business systems are difficult to interconnect, which restricts the comprehensive application of big data on equipment status and cannot support the development of new business forms such as integrated energy services and data operations. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a comprehensive energy data fusion method, device and equipment, which can complete the preliminary screening of heterogeneous data, business classification based on echo state network and data fusion based on semantic association feature mining, so as to solve the problem of low efficiency of comprehensive energy data fusion in the prior art.

[0005] In order to achieve the above-mentioned objectives, the first aspect of the present invention provides a comprehensive energy data fusion method, which includes: acquiring comprehensive energy data; the comprehensive energy data includes metering data, operating status data or monitoring data generated by energy sensing equipment; using a trained echo state network to determine the business type to which each comprehensive energy data belongs; extracting the data content of each comprehensive energy data according to the business type of each comprehensive energy data; and fusing the comprehensive energy data according to the semantic association feature corresponding to the data content.

[0006] Preferably, the fusion method also includes: after obtaining the comprehensive energy data, a step of removing abnormal data in the comprehensive energy data: mapping the comprehensive energy data to a coordinate system according to the attributes of the comprehensive energy data, and each comprehensive energy data corresponds to a data point; if the maximum value of the angle formed by a certain data point and the connecting line of the remaining data points is less than a set threshold; then the comprehensive energy data corresponding to the certain data point is removed as abnormal data.

[0007] Preferably, the echo state network includes: an input layer including multiple input neurons, a reserve pool including multiple reserve pool neurons, and an output layer including multiple output neurons; the multiple reserve pool neurons in the reserve pool form a ring topology; the trained echo state network includes: inputting samples of different business types into the echo state network to train the echo state network model; and using the trained echo state network as the trained echo state network.

[0008] Preferably, a trained echo state network is used to determine the business type to which the comprehensive energy data belongs, including: obtaining a business feature set of the comprehensive energy data, the business feature set including at least one business feature; inputting the business feature set into a trained echo state network to obtain a calculation result; and obtaining the business type corresponding to the comprehensive energy data according to the calculation result.

[0009] Preferably, the service characteristics include: size of service data, service duration, data packet length, and data packet arrival interval.

[0010] Preferably, the comprehensive energy data to be fused is subjected to data fusion according to the semantic association feature quantity corresponding to the data content, including: constructing the association feature distribution of the data content of the comprehensive energy data to be fused based on the random Gaussian observation matrix; constructing the objective function of big data semantic association feature fusion according to the association feature distribution; obtaining the data features corresponding to the data content according to the objective function; additively fusing the data content to obtain the data fusion result, and using the data features to update the data features of the data fusion result.

[0011] Preferably, the associated characteristic distribution of the data content of the comprehensive energy data to be fused is constructed based on the random Gaussian observation matrix, including:

[0012] Determine the data content is y i and j The equivalent semantic mapping relationship of the data content satisfies:

[0013]

[0014] where Φ ij for y i and j The random Gaussian measurement matrix between y i The associated feature distribution Θ(t) is:

[0015]

[0016] Among them, a mn Yes i The envelope amplitude value, g mn (t) is the statistical mean of the data, and n(t) is the interference noise.

[0017] Preferably, obtaining the data features corresponding to the data content according to the objective function includes: performing mean processing on the data content according to the objective function to obtain the common part of the data features; performing iterative optimization processing on the data content in combination with the objective function to obtain the independent part of the data features corresponding to the data content; and combining the common part of the data features and the independent part of the data features to form the data features corresponding to the data content.

[0018] In the second aspect of the present invention, a comprehensive energy data fusion device is also provided, including: a data acquisition module, used to acquire comprehensive energy data; the comprehensive energy data includes metering data, operating status data or monitoring data generated by energy sensing equipment; a business classification module, used to use a trained echo state network to determine the business type to which each comprehensive energy data belongs; a content extraction module, used to extract the data content of each comprehensive energy data according to the business type of each comprehensive energy data; and a fusion calculation module, used to fuse the comprehensive energy data according to the semantic association feature quantity corresponding to the data content.

[0019] In the third aspect of the present invention, there is also provided an integrated energy data fusion device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the aforementioned integrated energy data fusion method when executing the computer program.

[0020] Preferably, the fusion device is a chip.

[0021] Preferably, the memory and processor are integrated into a core board unit, and the fusion device further includes: an AC sampling unit for collecting and transmitting status parameters on the line; a security unit for providing keys to corresponding metering devices and corresponding distribution and transformation devices, and a local communication module for realizing data interaction of local devices.

[0022] Preferably, the fusion device also includes: a data storage unit for storing data from the core board unit; a display unit for displaying status parameters of the fusion device; a 4G\5G module for communicating with a remote master station; and at least one of the following interfaces: a remote control port, an RS232 / RS485 interface, an Ethernet interface, an RS-485 interface, a Bluetooth interface, a LoRa interface, an M-BUS interface, and a remote signal port.

[0023] In a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein instructions are stored in the storage medium, and when the instructions are executed on a computer, the computer executes the aforementioned integrated energy data fusion method.

[0024] A fifth aspect of the present invention provides a computer program product, comprising a computer program, which implements the aforementioned comprehensive energy data fusion method when executed by a processor.

[0025] The above technical solution has the following beneficial effects:

[0026] (1) Based on the simplified echo state network, the type perception of power business is realized, and the reserve pool structure is simplified into a ring topology composed of N neural unit nodes, which reduces the complexity of the reserve pool and solves the problem that the traditional echo state network algorithm is highly complex and difficult to use under the condition of limited edge computing resources.

[0027] (2) The fusion device or fusion equipment provided by the present invention is oriented to multiple business scenarios, can complete the collection of various energy information, and supports multiple communication methods, which is conducive to realizing the value of comprehensive energy and power big data.

[0028] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0030] Figure 1The schematic diagram of the implementation of the comprehensive energy data fusion method according to the embodiment of the present invention is schematically shown;

[0031] Figure 2 A schematic diagram of detecting and removing abnormal data according to an embodiment of the present invention is schematically shown;

[0032] Figure 3 A schematic diagram of a model of an echo state network according to an embodiment of the present invention is schematically shown;

[0033] Figure 4 A schematic diagram of the structure of a comprehensive energy data fusion device according to an embodiment of the present invention is shown;

[0034] Figure 5 The schematic diagram shows the structure of the comprehensive energy data fusion device according to the embodiment of the present invention. DETAILED DESCRIPTION

[0035] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0036] Figure 1 The following schematically shows an implementation diagram of the comprehensive energy data fusion method according to an embodiment of the present invention. Figure 1 As shown, this embodiment provides a comprehensive energy data fusion method, the fusion method comprising:

[0037] S01. Obtain comprehensive energy data; the comprehensive energy data includes metering data, operating status data or monitoring data generated by energy sensing equipment;

[0038] The data sources of the integrated energy system are very extensive. Data are collected through sensors on the usage of various energy sources, including the user's usage of electricity, water, gas, heat and other energy at a certain moment; monitoring of the operating status of various equipment, including monitoring data of equipment units such as wind, light, and storage, operation information of distributed energy of combined heat and power supply, and operation monitoring data of smart home appliances, smart furniture and other equipment.

[0039] S02. Use the trained echo state network to determine the business type to which each comprehensive energy data belongs; the echo state network (ESN) algorithm is a new type of reserve pool calculation model developed from a neural network. It uses a reserve pool composed of randomly sparsely connected neurons as a hidden layer, which greatly simplifies the deep learning process and ensures the global optimality and generalization ability of the algorithm model. The echo state network meets the real-time requirements of edge computing to a certain extent, and is an effective way to achieve real-time and high-speed computing on the edge side. However, the complexity of the original echo state network algorithm is still relatively high, and there is a problem of great practical difficulty when edge computing resources are limited. In this step, the echo state network is used to determine the business type, so as to improve the recognition efficiency under the constraints of fewer resources and real-time performance. Among them, the business types here can include: photovoltaic access, orderly charging, distribution monitoring, power quality management, reactive power compensation, fault location, etc.

[0040] S03. Extracting the data content of each piece of comprehensive energy data according to the business type of each piece of comprehensive energy data;

[0041] There are differences in the data format and data content of the integrated energy data of different business types. After identifying the business type, the data content can be extracted from the acquired integrated energy data according to the data format, providing a data basis for data fusion in the next step.

[0042] S04. Fusing the comprehensive energy data according to the semantic association feature quantities corresponding to the data contents.

[0043] Before data fusion, comprehensive energy data learns and extracts data features from different data acquisition device nodes to fully explore the data correlation structure and realize data fusion. Considering that the computing resources and storage resources of the execution terminal of data fusion are limited, this embodiment proposes a data feature extraction method of "semantic association feature mining". This method uses a semantic feature extractor to extract data features based on data cleaning and perception of multi-source heterogeneous data. The semantic association dimension feature quantity of the data is extracted in the reconstructed space, and the fused data is further optimized and the feature vector is provided to improve the efficiency of data fusion.

[0044] Through the above implementation, the efficiency of data fusion in energy data collection scenarios of multiple types of sensors can be improved. The echo state network and semantic association feature recognition in this implementation can achieve efficient recognition with lower system resources.

[0045] In an optional implementation provided by the present invention, the fusion method further includes: after obtaining the comprehensive energy data, a step of removing abnormal data in the comprehensive energy data: according to the attributes of the comprehensive energy data, the comprehensive energy data is mapped to a coordinate system, and each comprehensive energy data corresponds to a data point; if the maximum value of the angle formed by the connecting line of a certain data point and the remaining data points is less than a set threshold; then the comprehensive energy data corresponding to the certain data point is removed as abnormal data. Since data acquisition and status perception terminals are often in harsh environments, it is difficult to ensure the reliability and accuracy of the collected data. In addition, edge computing nodes are limited by size, computing and storage capabilities, and it is necessary to reduce the complexity of the algorithm to improve the feasibility of the algorithm. Therefore, it is necessary to clean the comprehensive energy multi-source heterogeneous data. First, a preliminary screening of the data is implemented to eliminate the noise information, null value information, error information, duplicate information and other content that are obviously present in the data; then the abnormal data is preliminarily screened. Figure 2 A schematic diagram of detecting and removing abnormal data according to an embodiment of the present invention is shown schematically. Figure 2 As shown. Assume that there are multiple pieces of comprehensive energy data collected, which are mapped one by one to multiple data points. Each data point is represented by a small circle, and the horizontal and vertical coordinates are different data attributes of the data. Traverse each data point and calculate the maximum angle among the angles formed by the data point and any other two data points. For normal data, such as B and C, the maximum angles α and β are relatively large angles, while for the abnormal point E, the maximum angle ω formed is relatively much smaller. This difference in angles can be used to effectively identify abnormal data. After the step of removing abnormal data, the set of remaining data points after preliminary screening is used as the input of business perception to determine the business type.

[0046] In an optional embodiment provided by the present invention, the echo state network includes: an input layer including multiple input neurons, a reserve pool including multiple reserve pool neurons, and an output layer including multiple output neurons; the multiple reserve pool neurons in the reserve pool form a ring topology; the trained echo state network includes: inputting samples of different business types into the echo state network to train the echo state network model; and using the trained echo state network as the trained echo state network. A traditional echo state network (ESN) usually includes: U input neurons, X reserve pool neurons, and Y output neurons. According to formula (1), the signal is transmitted from the input layer to the reserve pool neurons in the hidden layer. Where u(n) is the input layer function, x(n) is the hidden layer calculation result, and W in is the input connection matrix, W is the state matrix of the reserve pool, and f1(x) represents the activation function of the hidden layer.

[0047] x(n+1)=f1(Win u(n+1)+Wx(n)) (1)

[0048] Then the output function y(n) of the hidden layer neuron is:

[0049] y(n+1)=f2(W out u(n+1),x(n+1),y(n+1)) (2)

[0050] Among them, y(n) is the output layer function, f2(x) is the activation function of the output layer, and W out is the connection matrix from the reservoir to the output neural unit.

[0051] Figure 3 The schematic diagram of the model of the echo state network according to the embodiment of the present invention is schematically shown. The echo state network in this embodiment is an S-ESN model, that is, a simplified echo state network model. The S-ESN model mainly cuts the traditional reservoir calculation, while reducing the complexity of the reservoir in the echo state network, the reservoir structure is simplified to a ring topology composed of N neural unit nodes, and in order to maintain the accuracy of the reservoir operation while simplifying the original echo state network model, the nodes in the reservoir are generated by combining the dynamic equations with rich dynamic characteristics, as shown in formula (3).

[0052]

[0053] Where P is the average node generation rate, α is the excitation coefficient, δ is the node extinction rate, and τ is the node spacing. By integrating equation (3), we can get the neuron node x(n) in S-ESN:

[0054]

[0055] In an optional implementation manner provided by the present invention, a trained echo state network is used to determine the service type to which the comprehensive energy data belongs, including: obtaining a service feature set of the comprehensive energy data, the service feature set including at least one service feature; inputting the service feature set into the trained echo state network to obtain a calculation result; and obtaining the service type corresponding to the comprehensive energy data according to the calculation result. And the service features include: the size of the service data, the service duration, the data packet length, and the data packet arrival interval. The service perception method based on the S-ESN model is essentially a mapping of service features to service types, and its essence is to determine the classification and identification of service types based on service features. The extraction of service feature parameters is an important prerequisite for service identification, and how to select service feature parameters directly determines the performance of service identification. For each access service, the feature parameters of the received bidirectional data are extracted, including: the size of the service data, the service duration, the data packet length, and the data packet arrival interval. The feature parameters are normalized according to formula (5) to avoid overfitting, thereby obtaining a feature set C(i) describing the service, where D SIZE_MAX is the maximum business data counted, D TIME_MAX is the maximum service duration, P LENGTH_MAX is the maximum packet length, P INTERVAL_MAX The maximum interval between data packets arriving.

[0056]

[0057] The feature set C(i) of the service flow is input into the trained S-ESN, and the W out The classification to which the business belongs is obtained, thereby realizing the perception of various types of power business.

[0058] In the optional implementation manner provided by the present invention, the comprehensive energy data to be fused is subjected to data fusion according to the semantic association feature quantity corresponding to the data content, including: constructing the association feature distribution of the data content of the comprehensive energy data to be fused based on the random Gaussian observation matrix; constructing the objective function of the big data semantic association feature fusion according to the association feature distribution; obtaining the data features corresponding to the data content according to the objective function; additively fusing the data content to obtain the data fusion result, and simultaneously using the data features to update the data features of the data fusion result. And according to the objective function, performing mean processing on the data content to obtain the common part of the data features; iteratively optimizing the data content in combination with the objective function to obtain the independent part of the data features corresponding to the data content; and combining the common part of the data features and the independent part of the data features to form the data features corresponding to the data content. The specific calculation process is as follows:

[0059] First, we need to reconstruct the phase space of the cleaned and sensed data. Assume that any two sets of collected data samples are y i ,y j , where i,j∈(1,2,3…J), if the equivalent semantic mapping relationship satisfies:

[0060]

[0061] where Φ ij is the random Gaussian observation matrix between nodes i and j, then the collected data sample y to be mined i The associated feature distribution Θ(t) is:

[0062]

[0063] a mn Yes i The envelope amplitude value, g mn (t) is the statistical average of the data, and n(t) is the interference noise. The collected data is reconstructed in phase space, and the correlation dimension feature extraction method is used to construct the objective function of big data semantic correlation feature fusion, which is expressed as:

[0064]

[0065] Where p is the Euclidean distance between the reconstructed phase trajectories. For the semantic relevance feature fusion objective function, all collected data are averaged to obtain the common part of the fused data feature Θ C :

[0066]

[0067] For each set of sample data, the semantic correlation feature fusion objective function is combined to perform iterative optimization processing to obtain the independent part of the fused data feature Θ i The iterative form is:

[0068]

[0069] ω is the adaptive training weight, ω∈[0,1]; a ij is the embedding dimension of the phase space. In the data fusion stage, the intelligent terminal performs additive fusion on different collected data in turn, and records and stores the data features M(i) of each fusion step:

[0070] M(i)=M(i-1)∪{Θ C ,Θ i} (11)

[0071] Finally, the fused data obtained It is expressed as:

[0072]

[0073] Figure 4 The schematic diagram of the structure of the comprehensive energy data fusion device according to the embodiment of the present invention is shown as follows: Figure 4 A comprehensive energy data fusion device includes: a data acquisition module, used to acquire comprehensive energy data; the comprehensive energy data includes metering data, operation status data or monitoring data generated by energy sensing equipment; a business classification module, used to determine the business type of each comprehensive energy data by using a trained echo state network; a content extraction module, used to extract the data content of each comprehensive energy data according to the business type of each comprehensive energy data; and a fusion calculation module, used to fuse the comprehensive energy data according to the semantic association feature quantity corresponding to the data content.

[0074] The specific definition of each functional module in the above-mentioned integrated energy data fusion device can refer to the definition of the integrated energy data fusion method above, which will not be repeated here. Each module in the above-mentioned device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0075] In one embodiment of the present invention, a comprehensive energy data fusion device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the data collection method based on a mobile device as claimed in the preceding claim when executing the computer program. The processor here has the functions of numerical calculation and logical operation, and at least has a central processing unit CPU with data processing capability, a random access memory RAM, a read-only memory ROM, a variety of I / O ports and an interrupt system, etc. The processor includes a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and the aforementioned method is implemented by adjusting kernel parameters. The memory may include a non-permanent memory in a computer-readable medium, a random access memory (RAM) and / or a non-volatile memory, such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory includes at least one storage chip.

[0076] In some optional implementations provided by the present invention, the aforementioned comprehensive energy data fusion device capable of implementing the comprehensive energy data fusion method is a chip. Using a chip to implement the aforementioned comprehensive energy data fusion method has the advantages of high integration and easy deployment.

[0077] In some optional embodiments provided by the present invention, the memory and processor are integrated into the core board unit, and the fusion device also includes: an AC sampling unit for collecting and transmitting state parameters on the line; a security unit for providing keys to the corresponding metering equipment and the corresponding distribution and transformation equipment, and a local communication module for realizing data interaction of local devices. Specifically, the core board unit: mainly composed of a main control chip MCU, a power management chip, and a memory, and other hardware modules are connected to the core board unit. The core board unit is the main control part of the entire intelligent terminal. AC sampling unit: collects and transmits operating parameters, including various electrical quantities (voltage, current, power and other values ​​on the line), etc.; security unit: adopts a single-chip dual-key method, one key corresponds to the metering equipment, and the other corresponds to the distribution and transformation equipment; local communication module: adopts an HPLC high-speed power line carrier module or a micro-power wireless module to realize data interaction of local devices.

[0078] In some optional embodiments provided by the present invention, the fusion device further includes: a data storage unit for storing data from the core board unit; a display unit for displaying the status parameters of the fusion device; a 4G module for communicating with a remote master station; and at least one of the following interfaces: a remote control port, an RS232 / RS485 interface, an Ethernet interface, an RS-485 interface, a Bluetooth interface, a LoRa interface, an M-BUS interface, and a remote communication port. The functions of the above interfaces are as follows:

[0079] Remote control port: accept and execute remote control commands, used to control the opening and closing of switch equipment; RS232 / RS485 interface: used to access devices that support RS232 / RS485 communication mode, 232 / 485 is configurable; core board unit: mainly composed of main control chip MCU, power management chip, memory, and other hardware modules are connected to the core board, which is the main control part of the entire intelligent terminal. Local communication module: adopts HPLC high-speed power line carrier module or micro-power wireless module to realize data interaction of local devices; 4G\5G module: adopts 4G communication chip or 5G communication chip design to realize remote master station communication, compatible with 5G; Ethernet interface: adopts Ethernet chip design to realize local debugging communication; RS-485 interface: used to access devices that support 485 communication mode; Bluetooth interface: used to access devices that support Bluetooth communication mode; LoRa interface: used to access devices that support LoRa communication mode; M-BUS interface: used to support M-BUS bus device access; remote signal port: collect and transmit door node information.

[0080] Figure 5 The structure diagram of the comprehensive energy data fusion device according to the embodiment of the present invention is schematically shown. Figure 5As shown, the functions of each unit module in the figure are as described above and will not be repeated here. The integrated energy data fusion device provided in this embodiment supports multiple communication protocols, including 4G, 5G, power line carrier, RS485, LoRa, Bluetooth and other communication methods. The fusion device is suitable for a variety of application scenarios such as smart buildings, parks, industrial enterprises and user-side integrated energy. It can realize the access, analysis and processing of multi-source heterogeneous data, and better support the real-time analysis and intelligent processing of integrated energy local business through edge computing.

[0081] In one embodiment of the present invention, a computer-readable storage medium is also provided, in which instructions are stored. When the storage medium is run on a computer, the instructions, when executed by a processor, configure the processor to execute the above-mentioned integrated energy data fusion method.

[0082] In one embodiment of the present invention, a computer program product is provided, including a computer program, which implements the above-mentioned comprehensive energy data fusion method when executed by a processor.

[0083] Through the above implementation methods, the efficiency of comprehensive energy data fusion can be improved.

[0084] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0088] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0089] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0090] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0091] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0092] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A comprehensive energy data fusion method, characterized in that: The fusion method comprises: Obtaining comprehensive energy data; the comprehensive energy data includes metering data, operating status data or monitoring data generated by energy sensing devices; The trained echo state network is used to determine the business type to which each piece of comprehensive energy data belongs; Extract the data content of each piece of comprehensive energy data according to the business type of each piece of comprehensive energy data; The comprehensive energy data is subjected to data fusion according to the semantic association feature quantity corresponding to the data content of the comprehensive energy data, including: constructing the association feature distribution of the data content of the comprehensive energy data to be data fused based on the random Gaussian observation matrix; constructing the objective function of big data semantic association feature fusion according to the association feature distribution; obtaining the data feature corresponding to the data content according to the objective function; performing additive fusion on the data content to obtain the data fusion result, and using the data feature to update the data feature of the data fusion result; Among them, the associated characteristic distribution of the data content of the comprehensive energy data to be fused is constructed based on the random Gaussian observation matrix, including: Determine the data content is y i and j The equivalent semantic mapping relationship of the data content satisfies: where Φ ij for y i and j The random Gaussian measurement matrix between y i The associated feature distribution Θ(t) is: Among them, a mn Yes i The envelope amplitude value, g mn (t) is the statistical mean of the data, and n(t) is the interference noise.

2. The fusion method according to claim 1, characterized in that: The fusion method further comprises: after acquiring the comprehensive energy data, removing abnormal data in the comprehensive energy data: According to the attributes of the comprehensive energy data, the comprehensive energy data is mapped to a coordinate system, and each piece of comprehensive energy data corresponds to a data point; If the maximum value of the angle formed by a data point and the lines connecting the remaining data points is less than the set threshold, the comprehensive energy data corresponding to the data point will be removed as abnormal data.

3. The fusion method according to claim 1, characterized in that: The echo state network comprises: an input layer comprising a plurality of input neurons, a reservoir comprising a plurality of reservoir neurons, and an output layer comprising a plurality of output neurons; the plurality of reservoir neurons in the reservoir form a ring topology; The trained echo state network is obtained in the following manner: after samples of different service types are input into the echo state network, the echo state network model is trained, and the trained echo state network is used as the trained echo state network.

4. The fusion method according to claim 3, characterized in that: The trained echo state network is used to determine the business type to which the comprehensive energy data belongs, including: Acquire a service feature set of the comprehensive energy data, wherein the service feature set includes at least one service feature; Inputting the service feature set into a trained echo state network to obtain a calculation result; The business type corresponding to the comprehensive energy data is obtained according to the calculation result.

5. The fusion method according to claim 4, characterized in that: The service characteristics include: the size of service data, service duration, data packet length, and data packet arrival interval.

6. The fusion method according to claim 1, characterized in that: Obtaining data features corresponding to the data content according to the objective function includes: According to the objective function, the data content is processed by averaging to obtain the common part of the data features; Performing iterative optimization processing on the data content in combination with the objective function to obtain an independent part of the data feature corresponding to the data content; The common part of the data feature and the independent part of the data feature are used to form the data feature corresponding to the data content.

7. A comprehensive energy data fusion device, characterized in that: include: A data acquisition module, used to acquire comprehensive energy data; the comprehensive energy data includes metering data, operating status data or monitoring data generated by energy sensing devices; A business classification module is used to determine the business type to which each piece of comprehensive energy data belongs using a trained echo state network; A content extraction module, used to extract the data content of each piece of comprehensive energy data according to the business type of each piece of comprehensive energy data; as well as A fusion calculation module is used to perform data fusion on the comprehensive energy data according to the semantic association feature quantity corresponding to the data content of the comprehensive energy data, including: constructing the association feature distribution of the data content of the comprehensive energy data to be fused based on the random Gaussian observation matrix; constructing the objective function of big data semantic association feature fusion according to the association feature distribution; obtaining the data feature corresponding to the data content according to the objective function; performing additive fusion on the data content to obtain the data fusion result, and using the data feature to update the data feature of the data fusion result; Among them, the associated characteristic distribution of the data content of the comprehensive energy data to be fused is constructed based on the random Gaussian observation matrix, including: Determine the data content is y i and j The equivalent semantic mapping relationship of the data content satisfies: where Φ ij for y i and j The random Gaussian measurement matrix between y i The associated feature distribution Θ(t) is: Among them, a mn Yes i The envelope amplitude value, g mn (t) is the statistical mean of the data, and n(t) is the interference noise.

8. An integrated energy data fusion device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the comprehensive energy data fusion method described in any one of claims 1 to 6 are implemented.

9. The fusion device according to claim 8, characterized in that: The fusion device is a chip.

10. The fusion device according to claim 8, characterized in that: The memory and the processor are integrated into a core board unit, and the fusion device further includes: AC sampling unit, used to collect and transmit status parameters on the line; A security unit, used to provide a key to a corresponding metering device and a corresponding distribution transformer; and The local communication module is used to realize data interaction between local devices.

11. The fusion device according to claim 10, characterized in that: The fusion device also includes: A data storage unit, used for storing data from the core board unit; A display unit, used to display status parameters of the fusion device; 4G\5G module, used to achieve communication with the remote master station; and At least one of the following interfaces: Remote control port, RS232 / RS485 interface, Ethernet interface, RS-485 interface, Bluetooth interface, LoRa interface, M-BUS interface and remote signal port.

12. A computer-readable storage medium, characterized in that: The storage medium stores instructions, which, when executed on a computer, enable the computer to execute the comprehensive energy data fusion method described in any one of claims 1 to 6.

Images