Power distribution system
By using artificial intelligence models to predict load power demand and value data, multiple power allocation methods are provided, which solves the problem of insufficient power allocation priority ranking in existing technologies, and realizes the economic efficiency optimization of power allocation and the function of selling back surplus power.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- UNIVERSE AIOT TECHNOLOGY CO LTD
- Filing Date
- 2025-04-28
- Publication Date
- 2026-05-29
AI Technical Summary
The existing power distribution system cannot prioritize power allocation based on the importance of various electrical products and the value that may be lost due to power outages, resulting in all electrical products being affected equally during power outages or power rationing.
The system uses an artificial intelligence model to predict future power demand based on historical and real-time power usage data of the load, and combines this with value data from load operation to provide operation scripts for multiple power allocation methods for users to choose from in order to optimize power configuration.
It enables dynamic adjustment of power allocation priority based on the importance and value of the load, improves the economic efficiency of power configuration, and supports the resale of surplus power to other users or loads, providing the most economical power allocation scheme.
Smart Images

Figure CN224305416U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to a power distribution system, and more particularly to a power distribution system that can configure power based on historical usage data, real-time power usage status, and load operating value. Background Technology
[0002] The power distribution method after the user's electricity meter is to directly connect various loads, such as various electrical appliances. Therefore, during power outages or power restrictions, all electrical appliances will be affected by the same outage or power restriction conditions, and it is impossible to prioritize power distribution based on the importance of each electrical appliance and the value that may be lost due to the power outage.
[0003] Therefore, how to provide the best power allocation priority has become an important issue that needs to be addressed. Utility Model Content
[0004] One embodiment of this utility model discloses a power distribution system, comprising: a power supply for providing power; a distribution panel coupled to the power supply for receiving the power; and multiple loads coupled to the distribution panel, the distribution panel being used to generate control signals according to a first operation script to distribute the power to the multiple loads, wherein the distribution panel further comprises a control circuit, the control circuit predicting the power demand of each of the multiple loads based on historical power usage data and real-time power consumption data of each of the multiple loads, and generating the first operation script based on the operational value data of each of the multiple loads and the power demand of each of the multiple loads.
[0005] In some implementations, the power supply includes AC mains power and renewable energy.
[0006] In some embodiments, the distribution panel further includes: a main circuit coupled to the power supply for obtaining power from the power supply; and a plurality of branch circuits coupled to the main circuit and the plurality of loads, wherein the plurality of branch circuits distribute the power to the plurality of loads according to the control signal.
[0007] In some implementations, each of the plurality of branch circuits further includes a current and voltage sensing element for detecting the real-time power consumption data of each of the plurality of loads.
[0008] In some implementations, each of the multiple branch circuits further includes a switch, the control signal controlling the on / off state of the switch to distribute the power to the multiple loads.
[0009] In some implementations, each of the control circuit and the plurality of branch circuits further includes a communication module, through which the control circuit transmits the control signal to control the on / off state of the switch.
[0010] In some embodiments, the control circuit further includes: a data acquisition element that acquires the real-time power consumption data detected by each of the plurality of branch circuits via the communication module; a data analysis element coupled to the data acquisition element that generates a plurality of operation scripts based on the real-time power consumption data; and a control element coupled to the data analysis element that selects the first operation script among the plurality of operation scripts in response to a selection signal, wherein the control element controls the conduction state of the switch by transmitting the control signal through the communication module according to the first operation script.
[0011] In some implementations, each of the multiple operation scripts includes a corresponding power distribution mode and the economic value generated by distributing load power according to the corresponding power distribution mode.
[0012] In some implementations, the data analysis element further includes a processor and a memory, the memory storing a plurality of instructions, the processor executing different of the plurality of instructions to execute at least one artificial intelligence model to generate the plurality of operation scripts.
[0013] In some implementations, at least one artificial intelligence model includes: a load power demand model that predicts the power demand of each of the plurality of loads based on historical power usage data of each of the plurality of loads and the real-time power consumption data; and a value model that generates the plurality of operation scripts based on the value data of each of the plurality of loads during operation and the power demand of each of the plurality of loads.
[0014] This utility model's power distribution system uses artificial intelligence to predict the power demand of each load in the future time interval based on the load demand model generated by historical and real-time power consumption data of each load. It also provides a value model generated by historical value data of each load during operation, along with the power demand of each load in the future time interval, and offers multiple operation scripts with different power distribution methods and their corresponding economic values for users to choose from, so as to configure the power of the load according to the selected operation script. Attached Figure Description
[0015] To make the above and other objects, features, advantages and embodiments of this utility model more apparent and understandable, the accompanying drawings are described below:
[0016] Figure 1 The diagram shown is a schematic representation of a power distribution system according to a preferred embodiment of the present invention.
[0017] Figure 2 The diagram shown is a schematic representation of a data analysis element generating an operation script according to a preferred embodiment of the present invention.
[0018] Figure 3 The diagram shown is a schematic representation of the load demand model and value model according to a preferred embodiment of the present invention. Detailed Implementation
[0019] The following examples are described in detail with reference to the accompanying drawings. However, the provided examples are not intended to limit the scope of this invention, and the description of the structural operation is not intended to limit the order of execution. Any structure resulting from the recombination of elements and producing a device with equivalent functionality is within the scope of this invention. Furthermore, the illustrations are for illustrative purposes only and are not drawn to their original dimensions. For ease of understanding, the same or similar elements will be designated with the same symbols in the following description.
[0020] Unless otherwise specified, the terms used throughout the specification and claims generally have their ordinary meaning in the context of the art, the disclosure, and the specific content.
[0021] Furthermore, the terms "comprising," "including," "having," "containing," etc., used in this document are all open-ended terms, meaning "including but not limited to." Additionally, the term "and / or" as used in this document includes any one or more of the related listed items and all combinations thereof.
[0022] In this document, when an element is described as "connected," "coupled," or "electrically connected" to another element, the element may be directly connected, directly coupled, or directly electrically connected to the other element, or there may be an additional element between the two elements, while the element is indirectly connected, indirectly coupled, or indirectly electrically connected to the other element. However, when an element is described as "directly connected," "directly coupled," or "directly electrically connected" to another element, the two elements should be understood as having no additional element present. Furthermore, when an element is described as "wired" or "communicationally connected" to another element, the element may be indirectly connected to the other element via other elements for wired and / or wireless communication, or the element may be physically connected to the other element without needing other elements. Additionally, although terms such as "first," "second," etc., are used herein to describe different elements, these terms are only used to distinguish elements or operations described using the same technical terminology.
[0023] Since the power configuration after the user's electricity meter is directly connected to various loads, such as various electrical appliances, all electrical appliances will be affected by the same power outage or restriction conditions during a power outage or restriction. It is impossible to prioritize power allocation based on the importance of each electrical appliance and the potential value loss due to the power outage. Therefore, this utility model provides a power allocation system that uses artificial intelligence to predict the future power demand of each load over a certain time period based on a load demand model generated from historical and real-time power usage data, and a value model generated from historical value data of each load during operation. Combined with the future power demand of each load over a certain time period, it provides multiple operation scripts with different power allocation methods and their corresponding economic values for users to choose from, allowing for power allocation to the loads according to the selected operation script.
[0024] Figure 1 The diagram shown is a schematic representation of a power distribution system according to a preferred embodiment of the present invention. In some embodiments, the power distribution system 100 includes a distribution panel 110, a power supply 130, and multiple loads 141, 142, ..., 14n, where n is a positive integer. The distribution panel 110 is used to obtain power from the power supply 130 and distribute power to the multiple loads 141, 142, ..., 14n according to a power distribution method. In some embodiments, the power supply 130 includes AC mains power 131 and an auxiliary power supply 132. The auxiliary power supply 132 is renewable energy, including but not limited to power provided by wind power systems and solar photovoltaic systems, or power provided by energy storage devices. The loads 141, 142, ..., 14n are electrical appliances that use electricity, including but not limited to refrigeration equipment, lighting equipment, and life support equipment.
[0025] In some embodiments, the distribution panel 110 further includes a control circuit 150, a main circuit 120, and multiple branch circuits 121, 122, ..., 12n, where n is a positive integer. The control circuit 150 is used to generate the power distribution pattern. Each branch circuit 121, 122, ..., 12n has the same architecture, but is not limited thereto. The main circuit 120 is coupled to a power supply 130 and multiple branch circuits 121, 122, ..., 12n branching off from the main circuit 120. Accordingly, the current and voltage provided by the power supply 130 can be distributed through the main circuit 120 to the multiple branch circuits 121, 122, ..., 12n, thereby driving multiple loads 141, 142, ..., 14n connected to the multiple branch circuits 121, 122, ..., 12n. In some embodiments, each branch circuit 121, 122, ..., 12n includes at least a switch S and a current and voltage detection element D. The switch S is used to switch the coupling relationship between the corresponding branch circuits 121, 122, ..., 12n and the main circuit 120 according to the control signal generated by the control circuit 150 according to the power distribution method. The switch S can be, for example, an electromagnetic switch, but is not limited thereto. The current and voltage detection element D is used to detect the current and voltage provided by the corresponding branch circuits 121, 122, ..., 12n to the loads 141, 142, ..., 14n.
[0026] In some embodiments, the control circuit 150, each branch circuit 121, 122, ..., 12n, and the auxiliary power supply 132 each have a communication module M. Accordingly, the control circuit 150, each branch circuit 121, 122, ..., 12n, and the auxiliary power supply 132 can communicate with each other through their respective communication modules M. In some embodiments, the communication module M may be, for example, an infrared (IR) module, a Bluetooth module, a Wi-Fi module, etc., but this invention is not limited thereto.
[0027] In some embodiments, the control circuit 150 further includes a data acquisition element 151, a data analysis element 152, a control element 153, and a communication module M. The data acquisition element 151 can communicate with the communication module M of each branch circuit 121, 122, ..., 12n via the communication module M to acquire real-time voltage and current information of the load detected by the current and voltage detection elements D of each branch circuit 121, 122, ..., 12n. In some embodiments, the data acquisition element 151 acquires data every 5 minutes to calculate the power consumption of the load and perform real-time monitoring. Data analysis element 152 is coupled to data acquisition element 151. Based on the real-time voltage and current information, corresponding power consumption, and historical power usage information of each branch circuit 121, 122, ..., 12n provided to its respective load by the data acquisition element 151, it generates predicted power consumption for each load 141, 142, ..., 14n. Then, based on the economic value information of each load 141, 142, ..., 14n during operation, it generates multiple operation scripts. In some embodiments, historical power usage information includes, but is not limited to, the power consumption of each load and the peak or off-peak electricity price during usage time delays. In some embodiments, each operation script has a different load power allocation method and the corresponding economic benefit value under each power allocation method. Control element 153 is coupled to data analysis element 152. When the user selects one of the operation scripts, control element 153, according to this operation script, transmits control signals through communication module M to control the switches S of each branch circuit 121, 122, ..., 12n to switch the coupling relationship with the main circuit 120. In some implementations, if the user selects the power distribution method based on economic benefits, the operation script is to disconnect the power supply to the cooling equipment of load 141 and the lighting equipment of load 142, and maintain the power supply to the life support equipment of load 143. In this case, the control element 153, according to this operation script, transmits control signals through the communication module M to control the switches S of the branch circuits 121 and 122 to disconnect the coupling between load 141 and load 142 and the main circuit 120, and to control the switch S of the branch circuit 123 to maintain the coupling between load 143 and the main circuit 120, so that the power supply 130 continues to supply power to load 143.
[0028] In some embodiments, the data analysis element 152 includes at least one artificial intelligence model. This model generates multiple operation scripts based on real-time voltage and current information, power consumption information, and historical power usage information provided to each load by each branch circuit 121, 122, ..., 12n, as well as the economic value information of each load during operation. The data analysis element 152 includes a processor 1521 and a memory 1522. The memory 1522 also stores multiple instructions. The processor 1521 executes different of these instructions to execute the multiple operation scripts generated by the at least one artificial intelligence model. Notably, in the above embodiments, the data analysis element 152 is located in the distribution panel 110 as a ground-end element, executing the artificial intelligence model to generate multiple operation scripts and control the branch circuits 121, 122, ..., 12n. However, in other embodiments, the data analysis element 152 can also be located in the cloud, acquiring operation information from the distribution panel 110 via a communication mechanism, performing calculations in the cloud to generate multiple operation scripts, and then transmitting them back to the control circuit 150 to control the branch circuits 121, 122, ..., 12n.
[0029] Figure 2The diagram shows a schematic representation of an operation script generated by a data analysis element according to a preferred embodiment of the present invention. In some embodiments, at least one artificial intelligence model 200 includes a load power demand model 210 and a value model 220. The load power demand model 210 generates a power demand forecast for each load 141, 142, ..., 14n based on real-time voltage and current information, power consumption information, and historical power usage information provided to each load by each branch circuit 121, 122, ..., 12n. The power demand forecast for each load 141, 142, ..., 14n under the specified number of usage days 230 can be generated. The value model 220 generates an economic benefit value for each load 141, 142, ..., 14n based on historical data, including a value standard 240 for information such as the economic benefit value when the load is operating, the economic benefit value loss when the load is not operating, and the economic benefit value applicable to off-peak electricity prices during usage time delays. Then, based on the predicted power demand of future loads 141, 142, ..., 14n during batch time 250, and the corresponding economic benefit value of usage rules 260 for each load 141, 142, ..., 14n, multiple operation scripts 270 with different load power allocation methods can be generated. Each operation script has a different load power allocation method and a corresponding economic benefit value under each allocation method. Therefore, when the system provides the user with the operation script 270, the user can know in real time the corresponding power cost and the economic benefit value generated under each allocation method, and the user can then consider the operation script 270 with the best economic benefit for load power allocation. In some embodiments, through the optimal economic benefit power allocation provided by this invention, the user can sell excess power back to other users or loads to achieve optimal economic benefits. Therefore, this invention also provides a system and mechanism for clearing the sale or purchase of electricity, allowing users to know in advance that there is excess power available for sale to other users or loads when allocating power. In some implementations, the usage rules 260 for each load 141, 142, ..., 14n include, but are not limited to, rules such as minimum power requirements for load operation, minimum time requirements for load operation, or rules that the load must operate with priority.
[0030] Figure 3 The diagram shown is an architectural schematic of the load demand model and value model according to a preferred embodiment of the present invention. Figure 3The upper part shown is a schematic diagram of the architecture of the load power demand model 210. In some embodiments, the data used by the load power demand model 210 includes, but is not limited to, historical power usage information 211, load information 212, load power information 213, and other time-related information 214. In some embodiments, the historical power usage information 211 includes, but is not limited to, the load's past power consumption, corresponding usage time, and peak or off-peak electricity prices. Load information 212 includes, but is not limited to, information about the load itself, such as the load's power usage specifications and the purpose of its use. Load power information 213 includes, but is not limited to, the real-time voltage and current information and power consumption information provided by each branch circuit to its respective load. Other time-related information 214 includes, but is not limited to, activities that may require power at a future point in time. In some embodiments, by performing data cleaning 215 on the historical power usage information 211, load information 212, load power information 213, and other time-related information 214, and subsequent modeling training 216, a model 217 predicting the load's power demand at future points in time is finally generated.
[0031] on the other hand, Figure 3 The lower half shown is a schematic diagram of the value model 220 architecture. In some embodiments, the value model 220 includes a relatively stable lead time 221 and a parameter model 222 required for predicting power demand. Similarly, both undergo data cleaning 223, 225, model training 224, 226, etc., to establish a model 227 for lead processing and a model 228 for processing prediction parameters. Accordingly, the operation schedule 229 of each load can be set according to the economic benefit value, and after importing the model 217 that predicts the power demand of the load at future time points, an operation script 270 is generated. In some embodiments, because the real-time voltage and current information and power consumption information of the load change constantly, the power demand of the load predicted by the model 217 that predicts the power demand of the load at future time points will also change constantly. Therefore, this invention will dynamically change the operation script 270 according to the changes in the prediction results of the model 217 that predicts the power demand of the load at future time points, thereby providing users with the best economic benefit combination of load operation scripts 270 to choose from. In some implementations, the dynamically modified operation script 270 will also feed back to the pre-processing model 227 and the model 228 that processes prediction parameters, to calibrate the parameters of models 227 and 228 and improve the value model 220. In some implementations, the load demand model and the value model are obtained by performing multiple calculations using algorithms such as genetic algorithms, particle swarm optimization algorithms, and decision tree algorithms, supplemented by RPA (robotic process automation), and then taking the relatively optimal solution.
[0032] In summary, this utility model provides a power distribution system that uses artificial intelligence to predict the power demand of each load in the future time interval based on a load demand model generated from historical and real-time power consumption data of each load. It also provides a value model generated from historical value data of each load during operation, along with the power demand of each load in the future time interval. The system offers multiple operation scripts with different power distribution methods and their corresponding economic values for users to choose from, allowing for power configuration of the loads according to the selected operation script.
[0033] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the present invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be determined by the appended claims.
[0034] [Symbol Explanation]
[0035] To make the above and other objects, features, advantages and embodiments of this utility model more apparent and understandable, the appended symbols are explained as follows:
[0036] 100: Power Distribution System
[0037] 110: Distribution panel
[0038] 120: Main circuit
[0039] 121, 122, ..., 12n: Branch circuits
[0040] 130: Power supply
[0041] 131: AC mains power
[0042] 132: Auxiliary power supply
[0043] 141, 142, ..., 14n: Load
[0044] 150: Control Circuit
[0045] 151: Data Acquisition Component
[0046] 152: Data Analysis Components
[0047] 153: Control element
[0048] 1521: Processor
[0049] 1522: Memory
[0050] 200: Artificial Intelligence Model
[0051] 210: Load Power Demand Model
[0052] 211: Historical Electricity Usage Information
[0053] 212: Load Information
[0054] 213: Load Power Information
[0055] 214: Other Information
[0056] 215: Data Cleaning
[0057] 216: Modeling Training
[0058] 217: Model
[0059] 220: Value Model
[0060] 221: Lead Time
[0061] 222: Parametric Model
[0062] 223, 225: Data Cleaning
[0063] 224, 226: Modeling Training
[0064] 227: Model
[0065] 228: Model
[0066] 229: Operation Schedule
[0067] 230: Number of days used
[0068] 240: Value Standard
[0069] 250: Batch Time
[0070] 260: Usage Rules
[0071] 270: Operation Script
[0072] S: Switch
[0073] D: Current and voltage sensing element
[0074] M: Communication module.
Claims
1. A power distribution system, characterized in that, include: Power supply, providing electricity; A distribution panel, coupled to the power supply, is used to receive the power. as well as Multiple loads are coupled to the distribution panel, which generates control signals according to a first operation script to distribute power to the multiple loads. The distribution panel further includes a control circuit, which predicts the power demand of each of the plurality of loads based on the power usage history data and real-time power consumption data of each of the plurality of loads, and generates the first operation script based on the operating value data of each of the plurality of loads and the power demand of each of the plurality of loads.
2. The power distribution system according to claim 1, characterized in that, The power supply includes AC mains power and renewable energy.
3. The power distribution system according to claim 1, characterized in that, The distribution panel also includes: The main circuit is coupled to the power supply and obtains power from the power supply; and Multiple branch circuits are coupled to the main circuit and the multiple loads, wherein the multiple branch circuits distribute power to the multiple loads according to the control signal.
4. The power distribution system according to claim 3, characterized in that, Each of the plurality of branch circuits further includes a current and voltage detection element for detecting the real-time power consumption data of each of the plurality of loads.
5. The power distribution system according to claim 4, characterized in that, Each of the plurality of branch circuits further includes a switch, and the control signal controls the on state of the switch to distribute the power to the plurality of loads.
6. The power distribution system according to claim 5, characterized in that, Each of the control circuit and the plurality of branch circuits further includes a communication module, through which the control circuit transmits the control signal to control the conduction state of the switch.
7. The power distribution system according to claim 6, characterized in that, The control circuit also includes: The data acquisition element acquires the real-time power consumption data detected by each of the multiple branch circuits through the communication module; A data analysis element, coupled to the data acquisition element, generates multiple operation scripts based on the real-time power consumption data; and A control element, coupled to the data analysis element, selects the first operation script from the plurality of operation scripts in response to a selection signal. The control element controls the conduction state of the switch by transmitting the control signal through the communication module according to the first operation script.
8. The power distribution system according to claim 7, characterized in that, Each of the plurality of operation scripts includes a corresponding power distribution mode and the economic value generated by load power distribution according to the corresponding power distribution mode.
9. The power distribution system according to claim 7, characterized in that, The data analysis element further includes a processor and a memory, the memory storing multiple instructions, and the processor executing different of the multiple instructions to execute at least one artificial intelligence model to generate the multiple operation scripts.
10. The power distribution system according to claim 9, characterized in that, The at least one artificial intelligence model includes: A load power demand model predicts the power demand of each of the multiple loads based on historical power usage data and real-time power consumption data; and The value model generates multiple operation scripts based on the value data of each of the multiple loads during operation and the power demand of each of the multiple loads.