Electric Energy Meter Power Factor Management System

Abstract

The present invention relates to a power factor management system for an electric energy meter. The system includes: a content analysis device, configured to synchronously input each working current, each rated power, each resistance value corresponding to each load device currently managed by the electric energy meter, and multiple configuration parameters of the current electric energy meter into the factor analysis model, so as to obtain the lower limit value of the power factor of the current electric energy meter and the upper limit value of the power factor of the current electric energy meter; a change detection device, configured to trigger the content analysis device to execute the factor analysis model once when the number and / or type of each load device currently managed by the electric energy meter changes. Through this system, it is possible to directly analyze the power factor of the electric energy meter by introducing a factor analysis model with a customized structural design and a number of targeted basic data, so that it is not necessary to detect various indirect data by incorporating too many detection components, thereby improving the detection efficiency of the power factor of the electric energy meter.

Description

Technical Field

[0001] The present invention relates to the field of electric energy meters, and more specifically, to a power factor management system for electric energy meters. Background Art

[0002] An electric energy meter is an instrument used to measure electric energy, also known as a watt-hour meter, a fire meter, or a kilowatt-hour meter, which refers to an instrument for measuring various electrical quantities, commonly known as a watt-hour meter or a fire meter. The earliest electric energy meter was made based on the electrolysis principle in 1881. Although each such electric energy meter weighed dozens of kilograms, was very bulky and had no accuracy guarantee, it was still regarded as a major invention in the scientific and technological field at that time and was highly regarded and praised by people, and was quickly adopted in engineering. With the development of science and technology, the discovery and application of alternating current put forward new requirements for the development of electric energy meters. Through the efforts of scientists, the induction type electric energy meter was born. Due to a series of advantages such as simple structure, safe operation, low cost, durability, easy maintenance and mass production, the induction type electric energy meter has developed rapidly.

[0003] With the development of artificial intelligence technology, as one of the main types of power equipment, the requirements for the intelligent transformation of electric energy meters are getting higher and higher. The power equipment management party expects to directly obtain various indirect data at the electric energy meter end, rather than detecting various indirect data by incorporating too many detection components. For example, whenever there are changes in the quantity and / or type of each load device managed by the electric energy meter, the power equipment management party expects to directly obtain the power factor range of the electric energy meter limited by the lower power factor threshold and the upper power factor threshold through an intelligent analysis mode. Obviously, there is a lack of targeted intelligent data analysis solutions in the prior art. Summary of the Invention

[0004] To solve the technical problems in the prior art, the present invention provides a power factor management system for an electric energy meter. By introducing a step-by-step learning mechanism to perform each learning action on a deep feedforward neural network to perform step-by-step learning on the deep feedforward neural network, and outputting the deep feedforward neural network after the step-by-step learning as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electric energy meter. Thus, artificial intelligence models with different structures for performing power factor analysis of the electric energy meter are designed for different management states of different electric energy meters, and a number of basic data are specifically screened for the artificial intelligence model for performing power factor analysis of the electric energy meter, including each working current, each rated power, and each resistance value corresponding to each load device managed by the current electric energy meter, as well as the working frequency, operating voltage, and rated power of the current electric energy meter. This ensures the reliability and stability of the analysis results. A change detection device is also introduced to trigger the content analysis device to execute a factor analysis model only when it detects a change in the number and / or type of each load device managed by the current electric energy meter. Specifically, the change detection device determines whether there is a change in the number and / or type of each load device managed by the current electric energy meter based on the numerical change of the total resistance value of each load device managed by the current electric energy meter, thus avoiding misoperation of the factor analysis operation.

[0005] According to the present invention, there is provided a power factor management system for an electric energy meter, the system comprising:

[0006] A step-by-step learning mechanism for performing each learning action on a deep feedforward neural network to perform step-by-step learning on the deep feedforward neural network, and outputting the deep feedforward neural network after the step-by-step learning as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electric energy meter;

[0007] An information input mechanism for obtaining each working current, each rated power, and each resistance value corresponding to each load device managed by the current electric energy meter;

[0008] A parameter acquisition mechanism for outputting the working frequency, operating voltage, and rated power of the current electric energy meter as multiple configuration parameters of the current electric energy meter;

[0009] A content analysis device, respectively connected to the step-by-step learning mechanism, the information input mechanism, and the parameter acquisition mechanism, for synchronously inputting each working current, each rated power, and each resistance value corresponding to each load device managed by the current electric energy meter and multiple configuration parameters of the current electric energy meter into the factor analysis model to execute the factor analysis model, and obtaining the lower limit value of the power factor of the current electric energy meter and the upper limit value of the power factor of the current electric energy meter output by the factor analysis model;

[0010] A change detection device, connected to the content analysis device, is configured to trigger the content analysis device to execute a factor analysis model when it detects a change in the quantity and / or type of each load device managed by the current electricity meter.

[0011] Among them, the change detection device, connected to the content analysis device, is configured to trigger the content analysis device to execute a factor analysis model when it detects a change in the quantity and / or type of each load device managed by the current electricity meter, including: the change detection device determines whether there is a change in the quantity and / or type of each load device managed by the current electricity meter based on the numerical change of the total resistance value of each load device managed by the current electricity meter.

[0012] Among them, each learning action of the deep feedforward neural network is executed to perform step-by-step learning of the deep feedforward neural network, and the deep feedforward neural network after the step-by-step learning is output as a factor analysis model. The number of learning actions executed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electricity meter, including: using a positive correlation function to represent the numerical mapping relationship of the positive correlation between the number of learning actions executed by the deep feedforward neural network and the number of load devices managed by the current electricity meter.

[0013] Among them, each working current, each rated power, each resistance value corresponding to each load device managed by the current electricity meter, and multiple configuration parameters of the current electricity meter are synchronously input into the factor analysis model to execute the factor analysis model, and the lower limit value of the power factor of the current electricity meter and the upper limit value of the power factor of the current electricity meter output by the factor analysis model are obtained, including: the lower limit value of the power factor of the current electricity meter and the upper limit value of the power factor of the current electricity meter define the power factor range of the current electricity meter.

[0014] The present invention needs to have at least the following three important inventive points:

[0015] Inventive point A: Introduce a step-by-step learning mechanism to execute each learning action of the deep feedforward neural network to perform step-by-step learning of the deep feedforward neural network, and output the deep feedforward neural network after the step-by-step learning as a factor analysis model. The number of learning actions executed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electricity meter, so as to design artificial intelligence models with different structures for performing power factor analysis of electricity meters for different management states of different electricity meters.

[0016] Inventive Point B: A number of basic data are specifically screened for the artificial intelligence model for performing power factor analysis of the electricity meter, including the working currents, rated powers, and resistance values corresponding to each load device currently managed by the electricity meter, as well as the working frequency, operating voltage, and rated power of the current electricity meter, thereby ensuring the reliability and stability of the analysis results;

[0017] Inventive Point C: A change detection device is introduced to trigger the content analysis device to execute a factor analysis model when it detects a change in the quantity and / or type of each load device currently managed by the electricity meter. Specifically, the change detection device determines whether there is a change in the quantity and / or type of each load device currently managed by the electricity meter based on the numerical change in the total resistance value of each load device currently managed by the electricity meter, thereby avoiding misoperation of the factor analysis operation.

[0018] The power factor management system of the present invention operates stably and is intelligent in operation. Since a factor analysis model with a customized structural design and a number of specifically screened basic data can be introduced to directly analyze the power factor of the electricity meter, it is not necessary to detect various indirect data by incorporating too many detection components, thereby improving the detection efficiency of the power factor of the electricity meter. Brief Description of the Drawings

[0019] Those skilled in the art can better understand the numerous advantages of the present invention by referring to the drawings, wherein:

[0020] Figure 1 is a schematic structural diagram of the power factor management system of the electricity meter according to the primary embodiment of the present invention.

[0021] Figure 2 is a schematic structural diagram of the power factor management system of the electricity meter according to the secondary embodiment of the present invention.

[0022] Figure 3 is a schematic structural diagram of the power factor management system of the electricity meter according to the tertiary embodiment of the present invention. Detailed Description of the Embodiment

[0023] Figure 1 is a schematic structural diagram of the power factor management system of the electricity meter according to the primary embodiment of the present invention. The system includes:

[0024] A hierarchical learning mechanism for performing each learning action on the deep feedforward neural network to perform hierarchical learning on the deep feedforward neural network and output the deep feedforward neural network after the hierarchical learning is completed as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices currently managed by the electricity meter;

[0025] Specifically, a step-by-step learning mechanism is used to perform each learning action on a deep feedforward neural network to perform step-by-step learning on the deep feedforward neural network, and the deep feedforward neural network after the step-by-step learning is completed is output as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices currently managed by the electricity meter, including: selecting to use an ASIC chip to implement the step-by-step learning mechanism, which is used to perform each learning action on the deep feedforward neural network to perform step-by-step learning on the deep feedforward neural network, and the deep feedforward neural network after the step-by-step learning is completed is output as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices currently managed by the electricity meter;

[0026] An information entry mechanism is used to obtain each working current, each rated power, and each resistance value corresponding to each load device currently managed by the electricity meter;

[0027] A parameter acquisition mechanism is used to output the working frequency, operating voltage, and rated power of the current electricity meter as multiple configuration parameters of the current electricity meter;

[0028] A content analysis device is respectively connected to the step-by-step learning mechanism, the information entry mechanism, and the parameter acquisition mechanism, and is used to synchronously input each working current, each rated power, and each resistance value corresponding to each load device currently managed by the electricity meter and multiple configuration parameters of the current electricity meter into the factor analysis model to execute the factor analysis model, and obtain the lower limit value and upper limit value of the power factor of the current electricity meter output by the factor analysis model;

[0029] A change detection device is connected to the content analysis device, and is used to trigger the content analysis device to execute the factor analysis model once when it detects a change in the number and / or type of each load device currently managed by the electricity meter;

[0030] Among them, the change detection device is connected to the content analysis device, and is used to trigger the content analysis device to execute the factor analysis model once when it detects a change in the number and / or type of each load device currently managed by the electricity meter, including: the change detection device determines whether there is a change in the number and / or type of each load device currently managed by the electricity meter based on the numerical change of the total resistance value of each load device currently managed by the electricity meter;

[0031] Among them, each learning action is performed on the deep feedforward neural network to perform step-by-step learning of the deep feedforward neural network, and the deep feedforward neural network after the step-by-step learning is output as a factor analysis model. The number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electricity meter, including: using a positive correlation function to represent the numerical mapping relationship of the positive correlation between the number of learning actions performed by the deep feedforward neural network and the number of load devices managed by the current electricity meter;

[0032] Among them, each working current, each rated power, each resistance value corresponding to each load device managed by the current electricity meter, and multiple configuration parameters of the current electricity meter are synchronously input into the factor analysis model to execute the factor analysis model, and the lower limit value of the power factor of the current electricity meter and the upper limit value of the power factor of the current electricity meter output by the factor analysis model are obtained, including: the lower limit value of the power factor of the current electricity meter and the upper limit value of the power factor of the current electricity meter define the power factor range of the current electricity meter;

[0033] Among them, the change detection device determines whether the number and / or type of each load device managed by the current electricity meter has changed based on the numerical change of the total resistance value of each load device managed by the current electricity meter, including: when the numerical change of the total resistance value of each load device managed by the current electricity meter is greater than or equal to the set group value threshold, it is determined that the number and / or type of each load device managed by the current electricity meter has changed;

[0034] Among them, the change detection device determining whether the number and / or type of each load device managed by the current electricity meter has changed based on the numerical change of the total resistance value of each load device managed by the current electricity meter further includes: when the numerical change of the total resistance value of each load device managed by the current electricity meter is less than the set group value threshold, it is determined that the number and / or type of each load device managed by the current electricity meter has not changed.

[0035] Figure 2 It is a schematic structural diagram of an electricity meter power factor management system according to a secondary embodiment of the present invention.

[0036] Different from Figure 1 the electricity meter power factor management system in Figure 2 may further include the following components:

[0037] A wireless transceiver device, which is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and is used to receive the respective configuration information of the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times, so as to realize the real-time configuration of the current working modes of the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism;

[0038] Among them, the wireless transceiver device, which is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and is used to receive the respective configuration information of the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times, so as to realize the real-time configuration of the current working modes of the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism includes: the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through wireless communication links;

[0039] Among them, the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through time-division duplex communication links includes: the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism have different IP address data;

[0040] Among them, the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through wireless communication links includes: the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through time-division duplex communication links;

[0041] Among them, the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through wireless communication links includes: the wireless transceiver device is respectively connected to the content parsing device, the hierarchical learning mechanism, the information input mechanism, and the parameter acquisition mechanism through frequency-division duplex communication links.

[0042] Figure 3 It is a schematic structural diagram of an electric energy meter power factor management system according to a secondary embodiment of the present invention.

[0043] And Figure 1 different from Figure 3 the electric energy meter power factor management system in

[0044] The HDMI transmission device is respectively connected to the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and is used to provide data input operations and data output operations for the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times;

[0045] Among them, the HDMI transmission device is respectively connected to the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and is used to provide data input operations and data output operations for the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times, including: the HDMI transmission device includes multiple HDMI transmission interfaces, which are used to be respectively connected to the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and are used to provide data input operations and data output operations for the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times;

[0046] Among them, the HDMI transmission device includes multiple HDMI transmission interfaces, which are used to be respectively connected to the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and are used to provide data input operations and data output operations for the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times, including: the internal structures of the multiple HDMI transmission interfaces are the same;

[0047] And among them, the HDMI transmission device includes multiple HDMI transmission interfaces, which are used to be respectively connected to the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism, and are used to provide data input operations and data output operations for the content parsing device, the progressive learning mechanism, the information input mechanism, and the parameter acquisition mechanism at different times, including: the multiple HDMI transmission interfaces use the same pulse generating circuit.

[0048] In addition, in the power factor management system of the electric energy meter, using a positive correlation function to represent the numerical mapping relationship of the positive correlation between the number of learning actions performed by the deep feedforward neural network and the number of load devices currently managed by the electric energy meter includes: the number of load devices currently managed by the electric energy meter is the input parameter of the positive correlation function, and the number of learning actions performed by the deep feedforward neural network is the output parameter of the positive correlation function;

[0049] And among them, the number of load devices currently managed by the electricity meter is the input parameter of the positive correlation function, and the number of learning actions performed by the deep feedforward neural network is the output parameter of the positive correlation function, including: the fewer the number of load devices currently managed by the electricity meter, the fewer the number of learning actions performed by the deep feedforward neural network.

[0050] Those skilled in the art should understand that various modifications and changes can be made without departing from the scope and spirit of the present invention. Therefore, it should be understood that the above embodiments are for illustrative purposes only and are not limiting. Since the scope of the present invention is defined by the claims rather than the foregoing description, any changes and modifications that fall within the scope and boundaries of the claims or the equivalents of these scope and boundaries are subject to the claims.

Claims

1. A power factor management system for an electric energy meter, characterized in that: The system comprises: A step-by-step learning mechanism, used to perform each learning action on the deep feedforward neural network to perform step-by-step learning on the deep feedforward neural network, and output the deep feedforward neural network after the step-by-step learning as a factor analysis model, wherein the number of learning actions performed by the deep feedforward neural network is positively correlated with the number of load devices managed by the current electric energy meter; An information input mechanism is used to obtain the respective working currents, respective rated powers and respective resistance values ​​corresponding to the respective load devices currently managed by the electric energy meter; A parameter collection mechanism, used to output the current operating frequency, operating voltage and rated power of the electric energy meter as multiple configuration parameters of the current electric energy meter; The content analysis device is connected to the step-by-step learning mechanism, the information input mechanism and the parameter acquisition mechanism respectively, and is used to synchronously input the respective working currents, the respective rated powers and the respective resistance values ​​corresponding to the respective load devices managed by the current electric energy meter and the multiple configuration parameters of the current electric energy meter into the factor analysis model, so as to execute the factor analysis model and obtain the lower limit value of the power factor of the current electric energy meter and the upper limit value of the power factor of the current electric energy meter output by the factor analysis model; A change detection device, connected to the content analysis device, for triggering the content analysis device to execute a factor analysis model when detecting a change in the number and / or type of each load device managed by the current electric energy meter; The change detection device is connected to the content analysis device and is used to trigger the content analysis device to execute a factor analysis model when detecting that the number and / or type of each load device currently managed by the electric energy meter has changed. The change detection device determines whether the number and / or type of each load device currently managed by the electric energy meter has changed based on the value change of the total resistance value of each load device currently managed by the electric energy meter; When the value change of the total resistance value of each load device managed by the current electric energy meter is greater than or equal to the set group value threshold, it is determined that the number and / or type of each load device managed by the current electric energy meter has changed; Wherein, when the value change of the total resistance value of each load device managed by the current electric energy meter is less than the set group value threshold, it is determined that the number and / or type of each load device managed by the current electric energy meter has not changed; Wherein, a forward correlation function is used to represent a numerical mapping relationship of the positive correlation between the number of learning actions performed by the deep feedforward neural network and the number of load devices managed by the current electric energy meter; The number of load devices currently managed by the electric energy meter is the input parameter of the forward correlation function, and the number of learning actions performed by the deep feedforward neural network is the output parameter of the forward correlation function. The fewer the number of load devices currently managed by the electric energy meter, the fewer the number of learning actions performed by the deep feedforward neural network. The lower limit value of the current power factor of the electric energy meter and the upper limit value of the current power factor of the electric energy meter define the current power factor range of the electric energy meter.

2. The power factor management system of an electric energy meter according to claim 1, characterized in that: The system further comprises: A wireless transceiver device is connected to the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter collection mechanism, respectively, and is used to receive various configuration information of the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter collection mechanism in a time-sharing manner, so as to realize real-time configuration of the current working mode of the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter collection mechanism; Wherein, the wireless transceiver device is connected to the content analysis device, the step-by-step learning mechanism, the information input mechanism and the parameter collection mechanism respectively through wireless communication links.

3. The power factor management system of an electric energy meter as claimed in claim 2, characterized in that: The wireless transceiver device is connected to the content analysis device, the step-by-step learning mechanism, the information input mechanism and the parameter collection mechanism respectively through a time division duplex communication link.

4. The power factor management system of an electric energy meter as claimed in claim 3, characterized in that: The content analysis device, the step-by-step learning mechanism, the information input mechanism and the parameter collection mechanism have different IP address data.

5. The power factor management system of an electric energy meter as claimed in claim 3, characterized in that: The wireless transceiver device is connected to the content analysis device, the step-by-step learning mechanism, the information input mechanism and the parameter collection mechanism respectively through a frequency division duplex communication link.

6. The power factor management system of an electric energy meter according to claim 1, characterized in that: The system further comprises: An HDMI transmission device, connected to the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter acquisition mechanism, respectively, for providing data input operation and data output operation for the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter acquisition mechanism in a time-sharing manner; Among them, the HDMI transmission device includes multiple HDMI transmission interfaces, which are used to connect to the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter collection mechanism respectively, and are used to provide data input operations and data output operations for the content analysis device, the step-by-step learning mechanism, the information entry mechanism and the parameter collection mechanism in a time-sharing manner.

7. The power factor management system of an electric energy meter as claimed in claim 6, characterized in that: The internal structures of the multiple HDMI transmission interfaces are the same.

8. The power factor management system of an electric energy meter as claimed in claim 6, characterized in that: The multiple HDMI transmission interfaces use the same pulse generating circuit.

Images