Method and system for performing asset and energy management for electrical distribution system

Through sparse regression, the coefficient matrix and system representation are generated, and the machine learning model is trained to solve the prediction inaccurate problem caused by nonlinear correlation of variables in the power distribution system, and achieve high-accurate future state prediction and asset energy management.

CN118947038APending Publication Date: 2024-11-12ABB SPA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202280093102.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2022-03-28
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing distribution system assets and energy management systems fail to effectively consider the nonlinear correlation between variables, resulting in inaccurate predictions or errors, making it difficult to accurately predict complex system behaviors such as electrical loads.

Method used

A sparse regression method is used to generate a coefficient matrix, and the system representation is determined based on measurement data. It is used to train a machine learning model, considering the complex nonlinear relationships between variables, and adapting to changes in the nonlinear environment.

Benefits of technology

Improve the accuracy of future state prediction of power distribution systems, and can ideally describe the complex nonlinear relationships between variables, adapt to environmental changes, and achieve more accurate asset and energy management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118947038A_ABST
    Figure CN118947038A_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method of performing asset and energy management for an electrical distribution system (101). The method includes receiving measurement data of a plurality of input variables of the power distribution system (101). Further, the method includes generating, using sparse regression, a coefficient matrix for each output variable for asset and energy management based on the impact of the plurality of input variables on the corresponding output variable. The coefficient matrix includes one or more input variables from a plurality of input variables. Further, the method includes determining a plurality of system representations for each output variable based on the corresponding coefficient matrices. Each of the plurality of system representations indicates a relationship between one or more input variables and a corresponding output variable. Thereafter, the method comprises identifying a system representation from the plurality of system representations to train a machine learning model for predicting values of the output variables for performing asset and energy management for the power distribution system 101.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure generally relates to data analysis in power distribution systems. More specifically, the present disclosure relates to a system and method for performing asset and energy management for power distribution systems. Background Art

[0002] A distribution system is an electrical network that distributes electrical energy to end users for use. A distribution system may include multiple components, such as supply lines, transformers, switchgear, circuit breakers, switches, relays, fuses, cooling circuits, sockets, utilization components, etc. Components such as switchgear, circuit breakers, switches, relays, fuses, etc. protect other components of the distribution system, such as supply lines, transformers, utilization components, etc., by monitoring the conditions of other components. The main purpose of a distribution system is to provide a sustainable, affordable, and uninterrupted supply of electricity. To achieve this goal, asset management and energy management play a vital role. Typically, asset management includes monitoring the behavior of assets in a distribution system and restoring faults that cause power outages to end users. Asset management needs to maintain the functionality of assets and minimize asset failures to provide a sustainable and uninterrupted supply of electricity. For example, consider that there may be a short circuit on a distribution line. A distribution line may be connected to multiple assets of a distribution system. Asset management helps identify assets that cause a short circuit in a distribution line.

[0003] Energy management is required to plan both energy production and energy consumption. Data related to multiple variables of the power distribution system is used to perform asset and energy management. The power distribution system includes multiple constituent components, including electrical and mechanical components, and therefore includes multiple variables, such as electrical and mechanical variables. Typically, the correlation between these variables is complex because the variables exhibit nonlinearity, that is, there is no direct relationship between the variables, and the relationship between the variables is not predictable from a straight line. In a nonlinear relationship, the change in the output variable is not proportional to the change in the input variable.

[0004] Conventional systems that predict data to perform asset and energy management use machine learning to implement linear data modeling and analysis. However, these systems can result in inaccurate or error-prone predictions because they do not account for the complexities caused by nonlinearities and complex correlations between variables.

[0005] The information disclosed in this background of this disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art. Summary of the invention

[0006] In an embodiment, the present disclosure discloses a method for performing asset and energy management for a distribution system. The method includes receiving measurement data of multiple input variables of the distribution system from one or more sensors associated with the distribution system. Further, the method includes generating a coefficient matrix for each output variable from one or more output variables for asset and energy management. The coefficient matrix is ​​generated based on the influence of multiple input variables on corresponding output variables using sparse regression of the measurement data. The coefficient matrix includes one or more input variables from the multiple input variables. In addition, the method includes determining multiple system representations for each output variable based on the corresponding coefficient matrix. Each system representation in the multiple system representations indicates a relationship between one or more input variables and a corresponding output variable. Thereafter, the method includes identifying a system representation from the multiple system representations for each output variable. The system representation is used to train a machine learning model for predicting the value of the corresponding output variable to perform asset and energy management for the distribution system.

[0007] In an embodiment, the present disclosure discloses a computing system for performing asset and energy management for a power distribution system. The computing system includes a processor and a memory. The processor is configured to receive measurement data of multiple input variables of the power distribution system from one or more sensors associated with the power distribution system. Further, the processor is configured to generate a coefficient matrix for each output variable from one or more output variables for asset and energy management. The coefficient matrix is ​​generated based on the influence of multiple input variables on the corresponding output variables using sparse regression of the measurement data. The coefficient matrix includes one or more input variables from the multiple input variables. In addition, the processor is configured to determine multiple system representations for each output variable based on the corresponding coefficient matrix. Each system representation in the multiple system representations indicates a relationship between one or more input variables and a corresponding output variable. Thereafter, the processor is configured to identify a system representation from the multiple system representations for each output variable. The system representation is used to train a machine learning model for predicting the value of the corresponding output variable to perform asset and energy management for the power distribution system.

[0008] The foregoing summary is illustrative only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The novel features and characteristics of the present disclosure are set forth in the appended claims. However, the present disclosure itself, as well as the preferred mode of use, other objects and advantages thereof, will be best understood by referring to the following detailed description of illustrative embodiments when read in conjunction with the accompanying drawings. One or more embodiments will now be described, by way of example only, with reference to the accompanying drawings, wherein like reference numerals represent like elements, and wherein:

[0010] Figure 1 illustrates an exemplary environment for performing asset and energy management for a power distribution system according to some embodiments of the present disclosure;

[0011] Figure 2 illustrates a detailed diagram of a computing system for performing asset and energy management for a power distribution system according to some embodiments of the present disclosure;

[0012] Figure 3 An exemplary flow chart illustrating method steps for performing asset and energy management for a power distribution system according to some embodiments of the present disclosure is shown;

[0013] Figure 4 A block diagram of an exemplary computer system for performing asset and energy management for an electric distribution system is illustrated, according to some embodiments of the present disclosure.

[0014] Those skilled in the art will appreciate that any block diagram herein represents a conceptual diagram of an illustrative system that implements the principles of the present subject matter. Similarly, it will be appreciated that any flowchart, flow chart, state transition diagram, pseudocode, etc., represent various processes that can be substantially represented by a computer-readable medium and executed by a computer or processor (whether or not such a computer or processor is explicitly shown). DETAILED DESCRIPTION

[0015] In this document, the word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment or implementation of the subject matter described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0016] Although the present disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the accompanying drawings and will be described in detail below. However, it should be understood that it is not intended to limit the present disclosure to the particular form disclosed, but on the contrary, the present disclosure is intended to cover all modifications, equivalents and alternatives falling within the scope of the present disclosure.

[0017] The terms "comprises," "includes," or any other variation thereof are intended to cover a non-exclusive inclusion, such that a setup, device, or method that includes a list of components or steps includes not only those components or steps, but may also include other components or steps not expressly listed or inherent to such setup, device, or method. In other words, one or more elements in a system or device by "comprising" does not preclude the presence of other or additional elements in the system or device without further constraints.

[0018] Asset and energy management are the basic requirements of power distribution systems to ensure the availability of power assets, minimum failures, monitor energy consumption, etc. Generally, a power distribution system includes multiple components. These components may include multiple variables such as electrical and mechanical variables. There are complex correlations between the variables as the variables exhibit strong nonlinearity, parameter dependence, multi-scale phenomena, etc., making the final system behavior very complex. In a nonlinear relationship, the change in the output variable is not proportional to the change in any input variable. Further, multiple input variables may contribute to the output variable, making the prediction of the output variable a difficult process. For example, there may be multiple variables that contribute to the electrical load. Therefore, accurately predicting the electrical load is a difficult process. The input variables involve uncertainty and do not include a direct relationship with the output variable. Moreover, the output variable is characterized by nonlinearity and it may undergo rapid changes due to environmental changes, asset additions, demand patterns, customer maintenance requirements, macroeconomic changes.

[0019] Therefore, the present disclosure provides a method and system for performing asset and energy management for a distribution system. The present disclosure performs complex system analysis using nonlinear dynamics of measured data of the distribution system. The present disclosure provides high accuracy in predicting the future state of the distribution system from data because the present disclosure takes into account complex nonlinear correlations between variables through sparse regression. The present disclosure determines multiple system representations using changes in thresholding parameters and selection of variables for sparse regression. Further, the present disclosure identifies system representations that can ideally describe the complex nonlinearities between variables. Machine learning models can use system representations to accurately predict the future state of the distribution system to perform asset and energy management. The present disclosure can adapt to changing environments or complexities in nonlinear environments.

[0020] Figure 1 An exemplary environment 100 for performing asset and energy management for an electric distribution system (EDS) according to some embodiments of the present disclosure is illustrated. The exemplary environment 100 includes electric distribution systems 1011, 1012, 1013, ..., 101 N(collectively referred to as distribution system 101) and computing system 103. Distribution system 101 is an electrical network through which electrical energy is distributed to end users for use. Distribution system 101 includes multiple components, such as supply lines, transformers, switchgear, circuit breakers, switches, relays, fuses, cooling circuits, sockets, utilization components, etc. Components of distribution system 101 (such as switchgear, circuit breakers, switches, relays, fuses, etc.) protect other components of distribution system 101, such as supply lines, transformers, utilization components, etc., by monitoring other components. Distribution system 101 is connected to one or more sensors 1021, 1022, ..., 102 N (collectively referred to as one or more sensors 102) are associated. In an embodiment, one or more sensors 102 may be implemented within the power distribution system 101. For example, one or more sensors 102 may include a current sensor, a voltage sensor, etc. In another embodiment, one or more sensors 102 may be implemented outside the power distribution system 101. For example, one or more sensors 102 may include an environmental sensor, such as a temperature sensor, a humidity sensor, etc.

[0021] The computing system 103 may be configured to perform asset and energy management for the power distribution system 101. The computing system 103 may be a computing device such as a laptop, a desktop computer, a personal computer (PC), a notebook computer, a smart phone, a tablet computer, an e-book reader, a server, a network server, a cloud-based server, an edge computing device, a processing device, a router, etc. The computing system 103 may receive measurement data of multiple input variables of the power distribution system 101. The measurement data may be received from one or more sensors 102 associated with the power distribution system 101. For example, the measurement data may include values ​​of current, power, etc. Further, the computing system 103 may generate a coefficient matrix for each output variable from one or more output variables. One or more output variables may be used for asset and energy management. For example, one or more output variables may include power consumption, health status, cable and contact life, trip event classification of the power distribution system 101, etc.

[0022] The coefficient matrix can be generated based on the impact of multiple input variables on corresponding output variables using sparse regression on measured data. Sparse regression can be used to identify the most relevant input variables to predict the value of the output variable. The coefficient matrix includes one or more input variables from the multiple input variables. In addition, the computing system 103 can determine multiple system representations for each output variable based on the corresponding coefficient matrix. Each system representation in the multiple system representations indicates the relationship between one or more input variables and the corresponding output variable. Moreover, the computing system 103 can identify a system representation from the multiple system representations for each output variable. The system representation can be used to train a machine learning model for predicting the value of the corresponding output variable to perform asset and energy management for the distribution system 101.

[0023] In one implementation, the computing system 103 may be implemented as an edge computing platform. In such an embodiment, the computing system 103 may perform steps of identifying a system representation in an edge device by receiving measurement data from one or more sensors 102 associated with the power distribution system 101. The machine learning model may be implemented in the power distribution system 101 or a system associated with the power distribution system 101 (e.g., a system in an industrial plant including the power distribution system 101). In another embodiment, the machine learning model may be implemented in an edge device. In another implementation, the computing system 103 may be implemented in a cloud computing platform. In such an embodiment, the computing system 103 may perform steps of identifying a system representation in a cloud server by receiving measurement data from one or more sensors 102 associated with the power distribution system 101. The machine learning model may be implemented in the power distribution system 101 or a system or edge device associated with the power distribution system 101. In another implementation, the computing system 103 may be implemented in the power distribution system 101.

[0024] Figure 2 A detailed diagram of a computing system 103 for performing asset and energy management for a power distribution system 101 according to some embodiments of the present disclosure is illustrated. The computing system 103 includes an input / output (I / O) interface 201, a memory 202, and a processor 203. In some embodiments, the memory 202 may be communicatively coupled to the processor 203. The memory 202 stores instructions executable by the processor 203. The processor 203 may include at least one data processor to execute program components for executing requests generated by a user or a system. The memory 202 may be communicatively coupled to the processor 203. The memory 202 stores instructions executable by the processor 203, which, when executed, may cause the processor 203 to perform asset and energy management for the power distribution system 101.

[0025] In an embodiment, the memory 202 may include one or more modules 205 and computing data 204. The one or more modules 205 may be configured to perform the steps of the present disclosure using the computing data 204 to perform asset and energy management for the power distribution system 101. In an embodiment, each module 205 in the one or more modules 205 may be a hardware unit, which may be outside the memory 202 and coupled to the computing system 103. As used herein, the term module 205 refers to an application specific integrated circuit (ASIC), an electronic circuit, a field programmable gate array (FPGA), a programmable system on chip (PSoC), a combinational logic circuit, and / or other suitable components that provide the described functionality. When configured with the described functionality defined in the present disclosure, the one or more modules 205 will produce novel hardware. Further, the I / O interface 201 is coupled to the processor 203, and the input signal or / and the output signal are communicated through the processor 203. For example, the computing system 103 can communicate with one or more sensors 102 via the I / O interface 201.

[0026] In one implementation, the module 205 may include, for example, an input module 211, a matrix generation module 212, a system representation determination module 213, a system representation identification module 214, and other modules 215. It is to be understood that such above-mentioned modules 205 may be represented as a combination of a single module or different modules. In one implementation, the calculation data 204 may include, for example, input data 206, matrix data 207, system representation determination data 208, system representation identification data 209, and other data 210.

[0027] In an embodiment, the input module 211 may be configured to receive measurement data of multiple input variables of the power distribution system 101 from one or more sensors 102. The multiple input variables may include at least one of the following: electrical variables, mechanical variables, and environmental variables associated with the power distribution system 101. For example, the electrical variables may include variables such as current, voltage, power, power factor, harmonic distortion, energy, etc. Mechanical variables may include wear and tear curves, pressure, temperature, number of mechanical operations, etc. Environmental variables may include variables such as temperature, humidity, etc. In an example, consider that the power distribution system 101 includes components such as circuit breakers, fuses, voltage regulators, etc. Multiple input variables of the components may be received. In this case, the electrical variables of the circuit breaker may include arc current, short circuit events, contact resistance, etc. Mechanical variables may include contact pressure, contact wear, contact temperature and tear curves, number of mechanical operations, such as opening and closing of the circuit breaker, opening and closing distance travel, speed of moving contacts, etc. Environmental variables may include humidity, temperature, atmospheric pressure, etc. The measurement data of multiple input variables may include values ​​of multiple input variables. For example, the short-circuit current value of the circuit breaker may be 100KA. During the life of the circuit breaker, short-circuit events / faults may occur 3 to 4 times. In another example, during the life of the circuit breaker, the number of mechanical operations that occur may vary from 10,000 to 20,000. The input module 211 may receive measurement data of multiple input variables from one or more sensors 102 associated with the power distribution system 101. For example, one or more sensors 102 may include a current sensor, a voltage sensor, a temperature sensor, a pressure sensor, a humidity sensor, and the like. In an embodiment, one or more sensors 102 may be implemented in the power distribution system 101. For example, one or more sensors 102 may include a current sensor and a voltage sensor. In another embodiment, one or more sensors 102 may be implemented outside the power distribution system 101. For example, one or more sensors 102 may include a temperature sensor and a humidity sensor. In an example, the temperature sensor and the humidity sensor may be placed in the housing of the power distribution system 101. The measurement data of multiple input variables of the power distribution system 101 may be stored in the memory 202 as input data 206.

[0028] In an embodiment, the matrix generation module 212 can be configured to receive input data 206 from the input module 211. Further, the matrix generation module 212 can be configured to generate a coefficient matrix for each output variable from one or more output variables for asset and energy management. The one or more output variables for asset and energy management can include, for example, energy consumption, health of the distribution system 101, cable and contact life, trip event classification, etc. The coefficient generation module 212 can use sparse regression on the measured data to generate a coefficient matrix based on the impact of multiple input variables on the corresponding output variables. Sparse regression can identify the most relevant input variables to predict the value of the output variable.

[0029] The coefficients of sparse regression are sparse, that is, many variables are set to zero and only the most relevant variables are selected. In the present disclosure, sparse regression is used for multiple input variables that exhibit strong nonlinearity. Sparse regression identifies one or more input variables from multiple input variables based on the impact of multiple input variables on the corresponding output variable. For example, consider that one or more input variables identified from multiple input variables include current, pressure, voltage, temperature, arc current and resistance, and power of output variable energy consumption. Sparse regression can generate a coefficient matrix, which includes binary values ​​for each input variable in the multiple input variables. For example, the matrix generation module 212 can generate a coefficient matrix, wherein '0' values ​​are assigned to variables such as pressure, temperature, arc current, and the like, and '1' values ​​are assigned to one or more variables such as current, voltage, resistance, power, etc. The generated coefficient matrix can be stored in the memory 202 as matrix data 207.

[0030] In an embodiment, the system representation determination module 213 may be configured to receive matrix data 207 from the matrix generation module 212. Further, the system representation determination module 213 may be configured to determine multiple system representations for each output variable based on the corresponding coefficient matrix. Each system representation in the multiple system representations indicates the relationship between one or more input variables and the corresponding output variable. The system representation determination module 213 may determine one or more functions, wherein each function represents the relationship between the corresponding one or more input variables and the output variable. In an embodiment, the system representation determination module 213 may generate a candidate function matrix, which includes one or more functions associated with each of the corresponding one or more variables. Those skilled in the art will appreciate that one or more functions may be generated in any other form and are not limited to matrix form. The coefficient matrix combined with the candidate function matrix is ​​used to generate and determine multiple system representations. Each system representation in the multiple system representations includes one or more functions corresponding to each of the one or more input variables. In particular, the multiple system representations are a set of nonlinear control equations representing the power distribution system 101. Referring to the above example, the output variable may be energy consumption. The one or more variables may include current, voltage, resistance, power, etc. Functions may include, but are not limited to, linear functions, exponential functions, logarithmic functions, trigonometric functions, etc. The power distribution system 101 may be represented as a power system in the form of x'(t)=f(x(t)). For example, the system representation may be Y=six(x1)+cos 2 (2*x2)+sin 3 (5*x3).

[0031] In an example, the relationship between current and energy consumption can be defined by a linear function. The relationship between resistance and energy consumption can be defined by a linear function. In an example, the arc resistance of a circuit breaker may vary exponentially with temperature. Further, each of the multiple system representations can include a weighted value associated with each of the one or more input variables, the weight indicating the relationship between the output variable and the corresponding input variable. For example, consider that the function is a linear function. The weight may be "2". The system representation may be y=2x. In an embodiment, multiple system representations may include differential equations. An exemplary system representation is shown below:

[0032] x0=9.180+-0.119x0+0.026x1+-1.829x2+0.075x2^2............(1)

[0033] x1'=479.889+4.629x0+6.717x1+-60.062x2+-0.012x0^2+-0.011x0x1+-0.164x0

[0034] x2+0.015x1^2+0.116x1x2+1.793x2^2................(2)

[0035] x2'=3.076+-0.006x0+-0.409x2+0.012x2^2......................(33)

[0036] In an embodiment, the plurality of system representations are determined by variations in the plurality of input variables and thresholds associated with the plurality of input variables selected for sparse regression. The plurality of thresholds may indicate a degree of correlation between the plurality of input variables and the output variable. In an embodiment, one or more variables selected for sparse regression may be varied by execution to test different conditions, implementations, etc. The plurality of system representations determined for each output variable may be stored in the memory 202 as system representation determination data 208.

[0037] In an embodiment, the system representation identification module 214 may be configured to receive the system representation determination data 208 from the system representation determination module 213. Further, the system representation identification module 214 may be configured to identify the system representation from the plurality of system representations for each output variable. First, the system representation identification module 214 may determine the accuracy level of each of the plurality of system representations when determining the value of the output variable for the measurement data. Next, the system representation identification module 214 may identify the system representation with the greatest accuracy level from the plurality of system representations. The system representation that ideally describes the power distribution system 101 is identified. In an embodiment, the Akaike Information Criterion (AIC) may be used to identify the system representation from the plurality of system representations using the accuracy level as a criterion.

[0038] Those skilled in the art will appreciate that any known method other than the above method may be used to identify a system representation from a plurality of system representations. The system representation may be used to train a machine learning model to predict the value of a corresponding output variable using one or more machine learning techniques. For example, a machine learning model trained using a system representation may predict the value of a corresponding output variable to monitor asset life, cable and contact life, trip event classification, energy consumption, and the like. The identified system representation may be simulated for future states to predict the value of one or more output variables. The present disclosure identifies an ideal system representation that may be used to simulate future time behavior states of the distribution system 101 and may therefore be used to accurately predict system outputs for desired performance, such as asset and energy management for the distribution system 101.

[0039] Other data 210 may store data, including temporary data and temporary files, generated by one or more modules 205 for performing various functions of computing system 103. One or more modules 205 may also include other modules 215 to perform various miscellaneous functionality of computing system 103. Other data 210 may be stored in memory 202. It is to be appreciated that one or more modules 205 may be represented as a single module or a combination of different modules.

[0040] Figure 3 An exemplary flow chart illustrating method steps for performing asset and energy management for power distribution system 101 according to some embodiments of the present disclosure is shown. Figure 3 As illustrated, method 300 may include one or more steps. Method 300 may be described in the general context of computer-executable instructions. In general, computer-executable instructions may include routines, programs, objects, components, data structures, procedures, modules, and functions that perform specific tasks or implement specific abstract data types.

[0041] The order in which method 300 is described is not intended to be interpreted as limiting, and any number of the described method blocks can be combined in any order to implement the method. Additionally, individual blocks can be deleted from the method without departing from the scope of the subject matter described herein. In addition, the method can be implemented in any suitable hardware, software, firmware, or a combination thereof.

[0042] In step 301, computing system 103 may receive measurement data of a plurality of input variables of power distribution system 101. The plurality of input variables may include at least one of the following: an electrical variable, a mechanical variable, and an environmental variable associated with power distribution system 101. Computing system 103 may receive measurement data of the plurality of input variables from one or more sensors 102 associated with power distribution system 101.

[0043] In step 302, the computing system 103 may generate a coefficient matrix for each output variable from the one or more output variables for asset and energy management. The computing system 103 may generate the coefficient matrix based on the effects of the multiple input variables on the corresponding output variables using sparse regression on the measured data. The sparse regression may be used to generate the coefficient matrix based on selected thresholds of the multiple input variables and changes in the multiple input parameters.

[0044] In step 303, the computing system 103 may determine multiple system representations for each output variable based on the corresponding coefficient matrix. Each system representation in the multiple system representations indicates a relationship between one or more input variables and a corresponding output variable. The computing system 103 may determine one or more functions, wherein each function represents a relationship between the corresponding one or more input variables and the output variable. In an embodiment, the computing system 103 may generate a candidate function matrix, which includes one or more functions associated with each of the corresponding one or more variables. Those skilled in the art will appreciate that the one or more functions may be generated in any other form and are not limited to a matrix form. The coefficient matrix combined with the candidate function matrix is ​​used to generate and determine multiple system representations. Each system representation in the multiple system representations includes one or more functions corresponding to each of the one or more input variables. Each function represents a relationship between the corresponding one or more input variables and the output variable. The multiple system representations may include differential equations. In an embodiment, the multiple system representations are determined by changes in the multiple input variables and a threshold value, and the threshold value is related to the multiple input variables selected for sparse regression.

[0045] In step 304, computing system 103 may identify a system representation from the plurality of system representations for each output variable. Computing system 103 may determine a level of accuracy for each of the plurality of system representations when determining a value of the output variable for the measured data. Further, computing system 103 may identify a system representation having a maximum level of accuracy from the plurality of system representations. The system representations may be used to train a machine learning model to predict values ​​of corresponding output variables using one or more machine learning techniques. The machine learning model may be trained using the system representations. In particular, the machine learning model may simulate the system representations of future states to predict values ​​of one or more output variables.

[0046] Computer Systems

[0047] Figure 4A block diagram of an exemplary computer system 400 for implementing an embodiment consistent with the present disclosure is illustrated. In an embodiment, the computer system 400 may be used to implement the computing system 103. Thus, the computer system 400 may be used to perform asset and energy management for the power distribution system 101. In an embodiment, when the computer system 400 is implemented in an edge computing platform or a cloud computing platform, the computer system 400 may receive measurement data of multiple input variables from one or more sensors 102 via a communication network 409. In another embodiment, the computer system 400 may be implemented in the power distribution system 101. The computer system 400 may include a central processing unit 402 (also referred to as a "CPU" or "processor"). The processor 402 may include at least one data processor. The processor 402 may include a dedicated processing unit, such as an integrated system (bus) controller, a memory management control unit, a floating point unit, a graphics processing unit, a digital signal processing unit, and the like.

[0048] The processor 402 may be configured to communicate with one or more input / output (I / O) devices (not shown) via the I / O interface 401. The I / O interface 401 may employ communication protocols / methods such as, but not limited to, audio, analog, digital, mono, RCA, stereo, IEEE (Institute of Electrical and Electronics Engineers)-1394, serial bus, universal serial bus (USB), infrared, PS / 2, BNC, coaxial, component, composite, digital video interface (DVI), high-definition multimedia interface (HDMI), radio frequency (RF) antenna, super video, VGA, IEEE 802.n / b / g / n / x, Bluetooth, cellular (e.g., code division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), WiMax, etc.), etc.

[0049] Using the I / O interface 401, the computer system 400 can communicate with one or more I / O devices. For example, the input device 410 can be an antenna, a keyboard, a mouse, a joystick, an (infrared) remote control, a camera, a card reader, a fax machine, a dongle, a biometric reader, a microphone, a touch screen, a touch pad, a trackball, a stylus, a scanner, a storage device, a transceiver, a video device / source, a sensor, etc. The output device 411 can be a printer, a fax machine, a video display (e.g., a cathode ray tube (CRT), a liquid crystal display (LCD), a light emitting diode (LED), a plasma, a plasma display panel (PDP), an organic light emitting diode (OLED), etc.), an audio speaker, etc.

[0050] The processor 402 may be configured to communicate with a communication network 409 via a network interface 403. The network interface 403 may communicate with the communication network 409. The network interface 403 may use connection protocols including, but not limited to, direct connection, Ethernet (e.g., twisted pair 10 / 100 / 1000Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), Token Ring, IEEE802.11a / b / g / n / x, etc. The communication network 409 may include, but not limited to, direct interconnection, a local area network (LAN), a wide area network (WAN), a wireless network (e.g., using a wireless application protocol), the Internet, etc. The network interface 403 may use connection protocols including, but not limited to, direct connection, Ethernet (e.g., twisted pair 10 / 100 / 1000Base T), Transmission Control Protocol / Internet Protocol (TCP / IP), Token Ring, IEEE 802.11a / b / g / n / x, Bluetooth mesh, Zigbee, etc.

[0051] The communication network 409 includes, but is not limited to, direct interconnection, e-commerce networks, peer-to-peer (P2P) networks, local area networks (LANs), wide area networks (WANs), wireless networks (e.g., using wireless application protocols), the Internet, Wi-Fi, etc. The first network and the second network may be dedicated networks or shared networks, which represent associations of different types of networks that communicate with each other using various protocols, such as hypertext transfer protocol (HTTP), transmission control protocol / Internet protocol (TCP / IP), wireless application protocol (WAP), etc. Further, the first network and the second network may include various network devices, including routers, bridges, servers, computing devices, storage devices, etc.

[0052] In some embodiments, the processor 402 may be configured to communicate with the memory 405 (eg, Figure 4 The storage interface 404 can communicate with the memory 405 using connection protocols such as Serial Advanced Technology Attachment (SATA), Integrated Drive Electronics (IDE), IEEE-1394, Universal Serial Bus (USB), Fibre Channel, Small Computer System Interface (SCSI), etc., including but not limited to memory drives, removable disk drives, etc. The memory drive can also include a drum, a disk drive, a magneto-optical drive, an optical drive, a redundant array of independent disks (RAID), a solid-state memory device, a solid-state drive, etc.

[0053] Memory 405 may store a collection of program or database components, including but not limited to user interface 406, operating system 407, web browser 408, etc. In some embodiments, computer system 400 may store user / application data, such as data, variables, records, etc. described in the present disclosure. Such a database may be implemented as a fault-tolerant, relational, scalable, secure database, such as or

[0054] Operating system 407 may facilitate resource management and operation of computer system 400. Examples of operating systems include, but are not limited to, APPLE MACINTOSH R OS X, UNIX R , UNIX-like system distributions (such as BERKELEY SOFTWARE DISTRIBUTION TM (BSD), FREEBSD TM , NETBSD TM 、OPENBSD TM etc.), LINUX DISTRIBUTIONS TM (For example, RED HAT TM 、UBUNTU TM 、KUBUNTU TM etc.), IBM TM OS / 2, MICROSOFT TM WINDOWS TM (XP TM VISTA TM / 7 / 8, 10, etc.), APPLE R iOS TM 、GOOGLE R ANDROID TM ,BLACKBERRY R OS, etc.

[0055] In some embodiments, computer system 400 may implement a program component stored in web browser 408. Web browser 408 may be a hypertext viewing application such as MICROSOFT R INTERNET EXPLORER TM 、GOOGLE R CHROME TM0 、MOZILLA R FIREFOX TM 、APPLE R SAFARI TMetc. Secure web browsing can be provided using secure hypertext transfer protocol (HTTPS), secure socket layer (SSL), transport layer security (TLS), etc. The web browser 408 can utilize a web browser such as AJAX. TM , DHTML TM 、ADOBE R FLASH TM 、JAVASCRIPT TM , JAVA TM , application programming interface (API) and other facilities. In some embodiments, the computer system 400 can implement a program component stored in a mail server (not shown in the drawings). The mail server can be an Internet mail server such as Microsoft Exchange. The mail server can use a mail server such as ASP. TM 、ACTIVEX TM , ANSI TM C++ / C#、MICROSOFT R , .NET TM 、CGI SCRIPTS TM , JAVA TM 、JAVASCRIPT TM PERL TM , PHP TM 、PYTHON TM 、WEBOBJECTS TM Mail servers can use facilities such as Internet Message Access Protocol (IMAP), Messaging Application Programming Interface (MAPI), MICROSOFT R Exchange, Post Office Protocol (POP), Simple Mail Transfer Protocol (SMTP), and other communication protocols. In some embodiments, the computer system 400 can implement a program component stored in a mail client. The mail client (not shown in the figure) can be a mail viewing application, such as APPLE R MAIL TM MICROSOFT R ENTOURAGE TM MICROSOFT R OUTLOOK TM 、MOZILLA R THUNDERBIRD TM wait.

[0056] In addition, one or more computer-readable storage media can be used to implement embodiments consistent with the present disclosure. Computer-readable storage media refers to any type of physical memory that can store information or data that can be read by a processor. Therefore, a computer-readable storage medium can store instructions for one or more processors to execute, including instructions for causing (multiple) processors to execute steps or stages consistent with the embodiments described herein. The term "computer-readable medium" should be understood to include tangible items and does not include carrier waves and transient signals, that is, non-transient. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, non-volatile memory, hard disk drive, compact disk read-only memory (CD ROM), digital video disk (DVD), flash drive, disk, and any other known physical storage medium.

[0057] Embodiments of the present disclosure provide a method and system for performing asset and energy management for a distribution system. The present disclosure uses a nonlinear dynamics-based method to perform complex system analysis on measurement data associated with the distribution system. The present disclosure provides high accuracy in predicting the future state of the distribution system from the measurement data because the present disclosure takes into account the complex nonlinearity between variables through sparse regression. In particular, the present disclosure utilizes changes in thresholding parameters and the selection of variables for sparse regression to determine multiple system representations. Further, the present disclosure identifies an ideal system representation that describes the complex nonlinearity between variables and model system behavior. Machine learning models can use system representations to accurately predict the future state of the distribution system to perform asset and energy management. Further, the present disclosure can adapt to changing environments or complexities that typically occur in nonlinear environments by capturing real-time changes in the environment.

[0058] Unless expressly specified otherwise, the terms "one embodiment," "embodiments," "multiple embodiments," "the embodiment," "the multiple embodiments," "one or more embodiments," "some embodiments," and "an embodiment" refer to "one or more (but not all) embodiments of the present invention."

[0059] The terms "including", "comprising", "having" and variations thereof mean "including but not limited to", unless expressly specified otherwise.

[0060] An enumerated listing of items does not imply that any or all of the items are mutually exclusive, unless expressly specified otherwise. The terms "a," "an," and "the" mean "one or more," unless expressly specified otherwise.

[0061] The description of an embodiment with several components in communication with each other does not imply that all such components are required. Rather, various optional components are described to illustrate the various possible embodiments of the present invention.

[0062] When a single device or article is described herein, it is apparent that more than one device / article (whether or not they collaborate) may be used in place of a single device / article. Similarly, where more than one device or article is described herein (whether or not they collaborate), it is apparent that a single device / article may be used in place of more than one device / article, or a different number of devices / articles may be used in place of the number of devices or programs shown. The functionality and / or features of a device may alternatively be implemented by one or more other devices that are not explicitly described as having such functionality / features. Therefore, other embodiments of the present invention need not include the device itself.

[0063] Figure 3 The illustrated operations of show certain events that occur in a particular order. In alternative embodiments, certain operations may be performed, modified, or removed in a different order. Moreover, steps may be added to the above logic and still conform to the described embodiments. Further, the operations described herein may occur in sequence, or certain operations may be processed in parallel. Further, the operations may be performed by a single processing unit or a distributed processing unit.

[0064] Finally, the language used in the specification is selected primarily for readability and instructional purposes, and may not be selected to describe or constrain the subject matter of the invention. Therefore, the scope of the invention is not intended to be limited by this specific embodiment, but rather by any claims issued on an application based thereon. Therefore, the disclosure of the embodiments of the invention is intended to illustrate, but not to limit, the scope of the invention set forth in the following claims.

[0065] While various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope being indicated by the following claims.

[0066] Reference numerals

[0067]

[0068]

Claims

1. A method of performing asset and energy management of a power distribution system (101), the method comprising: Receiving measurement data of a plurality of input variables of the power distribution system (101) from one or more sensors (102) associated with the power distribution system (101); generating a coefficient matrix for each output variable from one or more output variables for asset and energy management based on the effects of the plurality of input variables on corresponding output variables using sparse regression on the measured data, wherein the coefficient matrix includes one or more input variables from the plurality of input variables; determining a plurality of system representations for each output variable based on the corresponding coefficient matrix, wherein each system representation of the plurality of system representations indicates a relationship between the one or more input variables and the corresponding output variable; as well as For each output variable, a system representation is identified from the plurality of system representations to train a machine learning model for predicting a value of the corresponding output variable for performing the asset and energy management for the power distribution system (101).

2. The method of claim 1, wherein the plurality of input variables comprises at least one of: electrical variables, mechanical variables, and environmental variables associated with the power distribution system (101).

3. The method of claim 1 , wherein each of the plurality of system representations comprises one or more functions corresponding to each of the one or more input variables, and wherein each function represents the relationship between the corresponding one or more input variables and the output variable.

4. The method of claim 1, wherein the plurality of system representations are determined by changes in the plurality of input variables and a threshold value associated with selecting the plurality of input variables for the sparse regression.

5. The method of claim 1 , wherein identifying the system representation from the plurality of system representations comprises: determining a level of accuracy for each of the plurality of system representations in determining the value of the output variable for the measurement data; as well as The system representation having a greatest level of accuracy is identified from the plurality of system representations.

6. The method of claim 1, wherein the value of the corresponding output variable is predicted using one or more machine learning techniques.

7. A computing system (103) for performing asset and energy management for a power distribution system (101), the computing system (103) comprising: Processor (203); as well as A memory (202), wherein the memory (202) stores processor-executable instructions, which, when executed, cause the processor (203): Receiving measurement data of a plurality of input variables of the power distribution system (101) from one or more sensors (102) associated with the power distribution system (101); generating a coefficient matrix for each output variable from one or more output variables for asset and energy management based on the effects of the plurality of input variables on corresponding output variables using sparse regression on the measured data, wherein the coefficient matrix includes one or more input variables from the plurality of input variables; determining a plurality of system representations for each output variable based on the corresponding coefficient matrix, wherein each system representation of the plurality of system representations indicates a relationship between the one or more input variables and the corresponding output variable; as well as A system representation is identified from the plurality of system representations for each output variable to train a machine learning model for predicting a value of the corresponding output variable for performing the asset and energy management for the power distribution system (101).

8. The computing system (103) of claim 7, wherein the plurality of input variables comprises at least one of: electrical variables, mechanical variables, and environmental variables associated with the power distribution system (101).

9. The computing system (103) of claim 7, wherein each of the plurality of system representations comprises one or more functions corresponding to each of the one or more input variables, and wherein each function represents the relationship between the corresponding one or more input variables and the output variable.

10. The computing system (103) of claim 7, wherein the processor (203) determines the plurality of system representations by changes in the plurality of input variables and a threshold value, the threshold value being associated with selecting the plurality of input variables for the sparse regression.

11. The computing system (103) of claim 7, wherein the processor (203) identifies the system representation from the plurality of system representations by: determining a level of accuracy for each of the plurality of system representations in determining the value of the output variable for the measurement data; and The system representation having a greatest level of accuracy is identified from the plurality of system representations.

12. The computing system (103) according to claim 7, implemented in one of the following: an edge computing platform, a cloud computing platform, and the power distribution system (101).

13. The computing system (103) of claim 7, wherein the processor (203) uses one or more machine learning techniques to predict the value of the corresponding output variable.