CN115877312A - Electric energy meter informatization evaluation calibration model based on station area electric energy conservation - Google Patents
Electric energy meter informatization evaluation calibration model based on station area electric energy conservation Download PDFInfo
- Publication number
- CN115877312A CN115877312A CN202211316897.XA CN202211316897A CN115877312A CN 115877312 A CN115877312 A CN 115877312A CN 202211316897 A CN202211316897 A CN 202211316897A CN 115877312 A CN115877312 A CN 115877312A
- Authority
- CN
- China
- Prior art keywords
- electric energy
- energy meter
- model
- optimization
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses an electric energy meter informatization evaluation calibration model based on station area electric energy conservation. Based on the station area electric energy conservation principle, hierarchical, accurate decoupling, measurement and collection and multilayer node cooperative calculation are realized, a non-convex sparse multi-target optimized mathematical model is constructed, a global adaptive optimization algorithm considering dynamic and static data is provided, and a station area line loss and electric energy meter evaluation calibration result is solved through cooperative coupling.
Description
Technical Field
The invention belongs to the technical field of power supply management, and particularly relates to an electric energy meter informatization evaluation calibration model based on station area electric energy conservation.
Background
At present, the electric energy meter state evaluation technology based on station area energy conservation is comprehensively popularized and applied in power enterprises, and the power enterprises deploy and apply an electric energy meter state evaluation model, so that a good effect is achieved in the aspect of improving the operation and maintenance level of the electric energy meter. However, the electric energy meter state evaluation technology still has the following problems: in the aspect of electric energy meter evaluation model construction, the existing model lacks quality prediction and early warning based on multi-dimensional data and lacks hierarchical energy loss accurate calculation. The existing evaluation model does not analyze and evaluate the state of the electric energy meter and predict the state trend of the electric energy meter by integrating history, reality performance and multi-dimensional data of cause and effect reasoning, only considers the reality state of the electric energy meter simply and in an isolated way, and has limitation on the scientificity of the evaluation result and guidance on future operation and use. Therefore, an actual physical model of the electric energy of the transformer area needs to be accurately established, and layered, graded, accurate decoupling measurement acquisition of key parameters and accurate measurement of multilayer node collaborative calculation parameters are achieved.
The non-convex sparse optimization problem refers to an optimization problem that a feasible solution has sparsity, and the problem generally has the characteristics of non-convex, non-continuous and the like and has NP complexity. At present, the work of constructing a non-convex sparse optimization model by students at home and abroad by utilizing the sparsity of errors of electric energy meters does not appear.
Disclosure of Invention
The invention aims to provide an electric energy meter informatization evaluation calibration model based on station area electric energy conservation, which is used for researching a layering, grading, accurate decoupling measurement and multilayer node cooperative computing technology based on a station area electric energy conservation principle to construct an electric energy meter informatization evaluation calibration physical model; the hybrid networking communication technology is researched, high-frequency data efficient compression transmission and clock automatic synchronization are achieved, and the real-time performance and the time scale accuracy of data high-frequency acquisition transmission are improved; constructing an electric energy meter evaluation calibration mathematical model based on a generating type confrontation network data quality enhancement and non-convex sparse multi-target optimization technology by taking a physical mechanism model as equality constraint; and researching a global self-adaptive optimization algorithm of prior data such as a first inspection error and the like, and realizing the cooperative coupling solution of the line loss of the transformer area and the evaluation and calibration result of the electric energy meter.
In order to realize the purpose of the invention, the technical scheme provided by the invention is as follows:
an electric energy meter informatization evaluation calibration model based on station electric energy conservation is constructed in the following mode:
step 1: the influence of the line loss measurement error of the transformer area on the calculation result of the model is reduced by utilizing a layered grading method based on the electric energy conservation principle and an accurate decoupling method of the line loss of the transformer area, and the uncertainty of the calculation result of the model is ensured to meet the requirements of a specific field; constructing an up-down error online dynamic calibration method by utilizing a multi-layer node collaborative computing technology of electric energy meter informatization evaluation calibration, and carrying out real-time evaluation and correction on the running states of the total nodes at different levels;
step 2: the method comprises the following steps of utilizing a hybrid networking technology based on HPLC to realize effective compression and efficient transmission of high-frequency data in the electric energy meter box; the NTB synchronization technology based on HPLC is utilized to realize automatic clock synchronization; the high efficiency and the accuracy of data acquisition of the electric energy meter are ensured;
and step 3: the method comprises the steps of simplifying and relaxing a multi-objective optimization model by using a non-convex sparse multi-objective optimization model construction method which takes electric energy meter errors as variables and takes data priorality, system stability and energy conservation as multiple optimization objectives, and using a mathematical technology to realize accurate estimation of the electric energy meter errors by using an efficient numerical optimization algorithm;
and 4, step 4: a multivariate optimization model construction method taking the line loss of the transformer area and the running error of the electric energy meter as variables is utilized, a local jump-out mechanism and a hyper-parameter adaptive selection technology are designed, a global adaptive algorithm based on an alternating direction multiplier method and a block coordinate descent method is designed, and the coupling solution of the line loss of the transformer area and the running error of the electric energy meter is realized.
Wherein, the step 1 specifically comprises the following steps:
step 1.1: by analyzing the typical structure of the existing low-voltage transformer area, a hierarchical structure model suitable for typical power utilization scenes of residential buildings and the like is described by taking a JP cabinet, a box transformer substation, a power distribution cabinet, a branch box and a metering box as basic elements;
step 1.2: based on different level models, establishing loss calculation models suitable for different levels by combining key measurement data based on time sequence load voltage, current, electric quantity and the like, converting an estimation problem that the sum difference between a total node and each branch node is line loss under a general condition into loss calculation models with specific physical meanings under different levels, and realizing accurate decoupling of loss;
step 1.3: through the multi-layer node collaborative calculation, the step-by-step transmission and tracing of the total node errors and the uncertainty under different levels are realized until the tail end node and the total node, the uncertain influence caused by the total node errors under different levels is further optimized, and the informatization evaluation and calibration of the electric energy meter are realized.
Wherein, the step 2 specifically comprises the following steps:
step 2.1: the method comprises the steps of utilizing a hybrid networking technology based on HPLC to realize the physical topology identification of a transformer area, and superposing characteristic current signals on the basis of HPLC to perform the physical topology identification in a hybrid networking mode; when the station transformer side management terminal calls the meter box side metering modules one by one, the meter box side metering modules sequentially generate characteristic current signals after receiving calling information; after receiving the signals, the branch side metering module records information and forwards characteristic signals to a management terminal, so as to construct a physical topology model;
step 2.2: by utilizing an NTB synchronization and data synchronization acquisition technology based on HPLC, firstly, a management terminal at a station transformer side serves as a CCO, metering modules at a station zone branch side and a meter box side serve as STA modules, broadcasting timing instructions are sent to the STA modules at the branch side and the meter box side through the CCO modules, accurate clock synchronization is carried out, the metering modules at the branch side and the meter box side are checked for roll calling one by one, and if clock second-level deviation exists, the clock deviation metering modules are re-calibrated; secondly, carrying out a research on a synchronous sampling technology of characteristic signals based on HPLC, when the voltage approaches over zero, superposing the voltage characteristic signals to realize information representation, and starting synchronous sampling after each stage of metering module receives the voltage characteristic signals in a voltage period;
step 2.3: the method comprises the steps that a high-frequency data compression transmission technology based on a hybrid networking technology is utilized to realize high-efficiency data transmission to a management terminal and a main station and carry out analysis and calculation, firstly, the high-frequency data compression transmission based on the hybrid networking technology is researched, a character string mode is replaced by numbers, a Huffman binary tree is constructed by using the occurrence frequency of characters, data with more occurrence times are arranged on the upper layer of the tree, and data with less occurrence times are arranged on the lower layer of the tree; coding is carried out on a path from the root node to each data, lossless compression is realized, and efficient transmission of the data is further realized according to a physical topological relation; and secondly, realizing the data sampling of the electrical parameter curve of the same time section based on a synchronous sampling technology.
Wherein, the step 3 specifically comprises the following steps:
step 3.1: the electric energy meter information evaluation calibration problem is accurately mapped by using a non-convex sparse multi-target optimization mathematical model construction method; based on the electric energy conservation principle of a transformer area, constructing a multi-objective optimization model which takes errors of an electric energy meter as variables and takes data priorality, system stability, energy conservation and the like as a plurality of optimization targets, and describing the data priorality targets by adopting a measurement function with non-convex and sparse mathematical characteristics; a measurement function with non-smooth mathematical characteristics is adopted to depict a system stability target; a measurement function with convex and smooth mathematical characteristics is adopted to depict an energy conservation target; finally, a non-convex sparse multi-target optimization model for accurately mapping the electric energy meter informatization evaluation calibration problem is constructed;
step 3.2: the original problem is reasonably simplified by using an effective relaxation method of a non-convex sparse multi-target optimization mathematical model of the electric energy meter; aiming at the non-convex, non-smooth and multi-target mathematical characteristics of the non-convex sparse multi-target optimization model, researching various convex methods of non-convex targets and corresponding convex approximation theory, and describing the error between the original problem solution and the relaxation problem solution; the method comprises the steps of describing the error between an original problem solution and a relaxation problem solution by utilizing a smoothing method of a non-smooth sparse target and a deep learning method of various smoothing model parameters; designing an autonomous learning strategy for parameters among various optimization targets based on various existing scaling methods, and developing a new scaling method;
step 3.3: the method comprises the steps that an efficient self-adaptive algorithm of a non-convex sparse multi-target optimization mathematical model of the electric energy meter is utilized to realize accurate identification of the out-of-tolerance electric energy meter; aiming at a special sparse structure of a model, a machine learning technology is combined, an efficient solving algorithm adaptive to a special sparse structure multi-target optimization problem is developed, the efficient solving algorithm is designed by using the calculation experience of the existing global optimization algorithm for reference, a local jump-out mechanism of a non-convex multi-target optimization problem global self-adaptive optimization algorithm is designed, and a better local optimal solution or global optimal solution is sought; the search direction and step length design of an approximation algorithm, a conjugate gradient algorithm and a confidence domain algorithm are utilized, so that the accuracy and the efficiency of the algorithm are improved; and further developing the theoretical analysis of algorithm convergence on the basis of the existing theoretical analysis results.
Wherein, the step 4 specifically comprises the following steps:
step 4.1: the method for estimating the line loss of the transformer area by utilizing dynamic and static data and a mathematical physical model to drive accurately describes the nonlinear function relationship among electric quantity, voltage, current, topology of the transformer area, line parameter dynamic and static data and line loss of the transformer area; calculating the voltage loss and the current loss of the current transformer area by adopting real-time dynamic data of voltage and current, static data of voltage impedance, current impedance and line parameters and a physical model of an electric power formula; by combining the calculated electric energy meter error data and utilizing a machine learning means, researching the nonlinear function relation among voltage loss, current loss, electric energy meter error, power supply and consumption difference, distribution room topological data and distribution room line loss, and realizing accurate estimation of the distribution room line loss;
step 4.2: constructing a multivariable coupling optimization model of the transformer area line loss and the electric energy meter operation error by utilizing a multivariable optimization model design method taking the transformer area line loss and the electric energy meter operation error as variables, combining a non-convex sparse optimization model of electric energy meter error estimation and a machine learning estimation model of the transformer area line loss;
step 4.3: solving the line loss of the transformer area and the running error of the electric energy meter by utilizing a separable alternative iteration algorithm and cooperative coupling; aiming at a multivariable optimization model, utilizing an augmented Lagrange function, a gradient function and a dual function thereof; decomposing the original problem into a plurality of optimization sub-problems by utilizing the separability of the objective function; constructing a distributed alternating direction multiplier method and a distributed block coordinate descent method, and solving optimization subproblems about line loss of a transformer area and errors of an electric energy meter in parallel; and constructing a self-adaptive shutdown criterion, accelerating the convergence speed of the algorithm, and analyzing the calculation complexity and the convergence of the algorithm.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention provides an electric energy meter evaluation and calibration method based on non-convex sparse multi-target optimization for the first time, considers the objective fact that the running states of most electric energy meters are good, creatively provides the method for reducing the solution volatility by using the sparsity of the running errors of the electric energy meters as a target function and using the difference between the current errors and the historical errors as the target function, and accordingly, a non-convex sparse multi-target optimization model is constructed, and the problem of electric energy meter evaluation and calibration is accurately mapped.
2. The invention firstly provides an optimization model for solving the coupling of the line loss of the transformer area and the operation error of the electric energy meter, provides a research idea for calculating the line loss of the transformer area by using dynamic and static data and a physical model, constructs a multivariable optimization model for coupling the error of the electric energy meter and the line loss of the transformer area, and designs a numerical iteration algorithm for gradually solving the line loss and the error alternately. The solution of the model not only helps to accurately estimate the error of the electric energy meter, but also helps to calculate the line loss of the distribution room.
Drawings
FIG. 1 is a schematic diagram of an electric energy meter informatization evaluation calibration model based on station electric energy conservation;
FIG. 2 is a schematic diagram of a non-convex sparse multi-target optimization mathematical model for electric energy meter evaluation and calibration in the invention;
FIG. 3 is a schematic diagram of a coupling optimization model of the transformer area line loss and the electric energy meter operation error.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the electric energy meter informatization evaluation calibration model based on station energy conservation according to the embodiment includes:
firstly, a hierarchical classification method based on the electric energy conservation principle and an accurate decoupling method of the line loss of the transformer area are utilized, the influence of the line loss measurement error of the transformer area on the model calculation result is reduced, and the uncertainty of the model calculation result is ensured to meet the requirements of a specific field. And constructing an up-down error on-line dynamic calibration method by utilizing a multi-layer node collaborative computing technology of electric energy meter information evaluation calibration, and evaluating and correcting the running states of the total nodes at different levels in real time.
It should be noted that the objective of the work at this stage is to construct a physical mechanism model under an electric energy meter informatization calibration hierarchical structure, eliminate or reduce the line loss rate absolute value and the fluctuation rate of the transformer area from the physical layer, and provide powerful theoretical support for the subsequent model solution. The method specifically comprises the following steps:
by analyzing the typical structure of the existing low-voltage transformer area, a hierarchical structure model suitable for typical power utilization scenes of residential buildings and the like is described by taking a JP cabinet, a box transformer substation, a power distribution cabinet, a branch (junction) box and a metering box as basic elements;
based on different level models, establishing loss calculation models suitable for different levels by combining key measurement data based on time sequence load voltage, current, electric quantity and the like, converting an estimation problem that the sum difference between a total node and each branch node is line loss under a general condition into loss calculation models with specific physical meanings under different levels, and realizing accurate decoupling of loss;
through the multi-layer node collaborative calculation, the step-by-step transmission and tracing of the total node errors and the uncertainty under different levels are realized until the tail end node and the total node, the uncertain influence caused by the total node errors under different levels is further optimized, and the informatization evaluation and calibration of the electric energy meter are realized.
Secondly, effective compression and efficient transmission of high-frequency data in the electric energy meter box are achieved by using a hybrid networking technology based on HPLC; and based on the NTB synchronization technology of HPLC, the automatic synchronization of clocks is realized; the high efficiency and the accuracy of the data acquisition of the electric energy meter are ensured.
It should be noted that the method specifically includes the following steps:
the physical topology identification of the transformer area is realized by using a hybrid networking technology based on HPLC, and the physical topology identification in a hybrid networking mode is carried out by superposing characteristic current signals on the basis of HPLC. When the station transformer side management terminal calls the meter box side metering modules one by one, the meter box side metering modules sequentially generate characteristic current signals after receiving calling information; and after receiving the signal, the branch side metering module records information and forwards the characteristic signal to the management terminal, so as to construct a physical topology model.
By utilizing an NTB synchronization and data synchronization acquisition technology based on HPLC, firstly, a management terminal at a station transformer side serves as a CCO, metering modules at a station zone branch side and a meter box side serve as STA modules, broadcasting timing instructions are sent to the STA modules at the branch side and the meter box side through the CCO modules to perform accurate clock synchronization, and the metering modules at the branch side and the meter box side are subjected to roll calling check one by one, if clock second-level deviation exists, the clock deviation metering modules are subjected to timing correction again. And secondly, carrying out a synchronous sampling technical study of characteristic signals based on HPLC, when the voltage approaches over zero, superposing the voltage characteristic signals to realize information representation, and starting synchronous sampling after each stage of metering module receives the voltage characteristic signals in a voltage period.
The method includes the steps that a high-frequency data compression transmission technology based on a hybrid networking technology is utilized, data are efficiently transmitted to a management terminal and a main station, analysis and calculation are carried out, firstly, the high-frequency data compression transmission based on the hybrid networking technology is researched, a character string mode is replaced by numbers, a Huffman binary tree is constructed by using the occurrence frequency of characters, data with more occurrence times are on the upper layer of the tree, and data with less occurrence times are on the lower layer of the tree. And coding is carried out on a path from the root node to each data, lossless compression is realized, and efficient transmission of the data is further realized according to the physical topological relation. And secondly, realizing the data sampling of the electrical parameter curve of the same time section based on a synchronous sampling technology.
Thirdly, as shown in fig. 2, by using a non-convex sparse multi-objective optimization model construction method which takes the electric energy meter error as a variable and takes data prioriness, system stability and energy conservation as a plurality of optimization targets, and by using mathematical techniques such as convex, smooth and standard quantization, the multi-objective optimization model is reasonably simplified and relaxed, and an efficient numerical optimization algorithm is designed to realize accurate estimation of the electric energy meter error.
Specifically, the following are included:
firstly, a non-convex sparse multi-target optimization mathematical model construction method of the electric energy meter is utilized to accurately map the problem of informatization evaluation and calibration of the electric energy meter. Based on the electric energy conservation principle of a transformer area, constructing a multi-objective optimization model which takes electric energy meter errors as variables and takes data priori, system stability, energy conservation and the like as a plurality of optimization targets, and describing the data priori targets by adopting measurement functions with non-convex and sparse mathematical characteristics such as L0 norm, lp norm and the like of the electric energy meter errors; the method comprises the following steps of describing a system stability target by using a measurement function with non-smooth mathematical characteristics, such as an absolute value function, an SCAD function and the like between a current error and a historical error; the energy conservation target is characterized by measuring functions with convex and smooth mathematical characteristics, such as quadratic functions, polynomial functions and the like of the difference between the power supply quantity and the power consumption quantity; and finally, constructing a non-convex sparse multi-target optimization model for accurately mapping the electric energy meter informatization evaluation calibration problem.
Secondly, an effective relaxation method of the electric energy meter non-convex sparse multi-target optimization mathematical model is utilized to reasonably simplify the original problems. Aiming at the mathematical characteristics of non-convex, non-smooth, multi-target and the like of the non-convex sparse multi-target optimization model, researching various convex methods of the non-convex target and a corresponding convex approximation theory, and describing the error between the original problem solution and the relaxation problem solution; the method comprises the steps of describing the error between an original problem solution and a relaxation problem solution by utilizing a smoothing method of a non-smooth sparse target and a deep learning method of various smoothing model parameters; based on various existing scaling methods, an autonomous learning strategy for optimizing parameters among targets is designed, and a new scaling method is developed.
And finally, realizing accurate identification of the out-of-tolerance electric energy meter by using a high-efficiency self-adaptive algorithm of the non-convex sparse multi-target optimization mathematical model of the electric energy meter. Aiming at a special sparse structure of a model, a machine learning technology is combined, an efficient solving algorithm adaptive to a special sparse structure multi-target optimization problem is developed, and the method comprises the steps of designing a local jumping-out mechanism of a non-convex multi-target optimization problem global self-adaptive optimization algorithm by using the calculation experience of the existing global optimization algorithm for reference, and seeking a more optimal local optimal solution or global optimal solution; search direction and step length design of methods such as an approximation algorithm, a conjugate gradient algorithm, a confidence domain algorithm and the like are researched, so that the algorithm precision and efficiency are improved; and further developing the theoretical analysis of algorithm convergence on the basis of the existing theoretical analysis results.
Fourthly, as shown in fig. 3, by using a multivariate optimization model construction method taking the line loss of the transformer area and the operation error of the electric energy meter as variables, the techniques of a local jump-out mechanism, hyper-parameter adaptive selection and the like are researched, a global adaptive algorithm based on an alternating direction multiplier method and a block coordinate descent method is designed, and the coupling solution of the line loss of the transformer area and the operation error of the electric energy meter is realized.
Specifically, the following are included:
firstly, accurately describing a nonlinear function relation between dynamic and static data such as electric quantity, voltage, current, transformer area topology, line parameters and the like and transformer area line loss by using a transformer area line loss estimation method driven by dynamic and static data and a mathematical physical model; calculating the voltage loss and the current loss of the current transformer area by using real-time dynamic data such as voltage, current and the like and static data such as voltage impedance, current impedance, line parameters and the like and by using physical models such as an electric power formula and the like; by combining the electric energy meter error data calculated by the subject, a machine learning means is utilized to research the nonlinear function relationship between the data such as voltage loss, current loss, electric energy meter error, power supply and consumption difference, distribution room topology and the like and distribution room line loss, so that accurate estimation of the distribution room line loss is realized.
Secondly, a design scheme of a multivariable optimization model with the line loss of the transformer area and the running error of the electric energy meter as variables is utilized. And constructing a multivariable coupling optimization model of the line loss of the transformer area and the operation error of the electric energy meter by combining a non-convex sparse optimization model of the error estimation of the electric energy meter and a machine learning estimation model of the line loss of the transformer area.
And finally, solving the line loss of the transformer area and the running error of the electric energy meter by utilizing separable alternative iterative algorithms such as an alternative direction multiplier method, a block coordinate descent method and the like in a cooperative coupling manner. Researching an augmented Lagrange function, a gradient function and a dual function of the multivariate optimization model; decomposing the original problem into a plurality of optimization sub-problems by utilizing the separability of the objective function; constructing a distributed alternating direction multiplier method and a distributed block coordinate descent method, and solving optimization subproblems about line loss of a transformer area and errors of an electric energy meter in parallel; and constructing a self-adaptive shutdown criterion, accelerating the convergence speed of the algorithm, and analyzing the computational complexity and the convergence of the algorithm.
Finally, it should be noted that: the foregoing examples are provided for illustration and description of the invention only and are not intended to limit the invention to the scope of the described examples. Furthermore, it will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that many variations and modifications may be made in accordance with the teachings of the present invention, all of which fall within the scope of the invention as claimed.
Claims (5)
1. An electric energy meter informatization evaluation calibration model based on station electric energy conservation is characterized by being constructed in the following mode:
step 1: the influence of the line loss measurement error of the transformer area on the calculation result of the model is reduced by utilizing a layered grading method based on the electric energy conservation principle and an accurate decoupling method of the line loss of the transformer area, and the uncertainty of the calculation result of the model is ensured to meet the requirements of a specific field; constructing an up-down error online dynamic calibration method by utilizing a multi-layer node collaborative computing technology of electric energy meter informatization evaluation calibration, and carrying out real-time evaluation and correction on the running states of the total nodes at different levels;
step 2: the method comprises the following steps of utilizing a hybrid networking technology based on HPLC (high performance liquid chromatography), and realizing effective compression and efficient transmission of high-frequency data in the electric energy meter box; the NTB synchronization technology based on HPLC is utilized to realize automatic clock synchronization; the high efficiency and the accuracy of data acquisition of the electric energy meter are ensured;
and step 3: the method comprises the steps of simplifying and relaxing a multi-objective optimization model by using a non-convex sparse multi-objective optimization model construction method which takes electric energy meter errors as variables and takes data priorality, system stability and energy conservation as multiple optimization objectives, and using a mathematical technology to realize accurate estimation of the electric energy meter errors by using an efficient numerical optimization algorithm;
and 4, step 4: a multivariate optimization model construction method taking the line loss of the transformer area and the running error of the electric energy meter as variables is utilized, a local jump-out mechanism and a hyper-parameter adaptive selection technology are designed, a global adaptive algorithm based on an alternating direction multiplier method and a block coordinate descent method is designed, and the coupling solution of the line loss of the transformer area and the running error of the electric energy meter is realized.
2. The electric energy meter informatization evaluation calibration model based on station area electric energy conservation according to claim 1, wherein the step 1 specifically comprises the following steps:
step 1.1: by analyzing the typical structure of the existing low-voltage transformer area, a hierarchical structure model suitable for typical power utilization scenes of residential buildings and the like is described by taking a JP cabinet, a box transformer substation, a power distribution cabinet, a branch box and a metering box as basic elements;
step 1.2: based on different level models, establishing loss calculation models suitable for different levels by combining key measurement data based on time sequence load voltage, current, electric quantity and the like, converting an estimation problem that the sum difference between a total node and each branch node is line loss under a general condition into loss calculation models with specific physical meanings under different levels, and realizing accurate decoupling of loss;
step 1.3: through the multi-layer node collaborative calculation, the step-by-step transmission and tracing of the total node errors and the uncertainty under different levels are realized until the tail end node and the total node, the uncertain influence caused by the total node errors under different levels is further optimized, and the informatization evaluation and calibration of the electric energy meter are realized.
3. The electric energy meter informatization evaluation calibration model based on station area electric energy conservation according to claim 1, wherein the step 2 specifically comprises the following steps:
step 2.1: the method comprises the steps of utilizing a hybrid networking technology based on HPLC to realize the physical topology identification of a transformer area, and superposing characteristic current signals on the basis of HPLC to perform the physical topology identification in a hybrid networking mode; when the station transformer side management terminal calls the meter box side metering modules one by one, the meter box side metering modules sequentially generate characteristic current signals after receiving calling information; after receiving the signals, the branch side metering module records information and forwards characteristic signals to a management terminal, so as to construct a physical topology model;
step 2.2: by utilizing an NTB synchronization and data synchronization acquisition technology based on HPLC, firstly, a management terminal at a station transformer side serves as a CCO, metering modules at a station zone branch side and a meter box side serve as STA modules, broadcasting timing instructions are sent to the STA modules at the branch side and the meter box side through the CCO modules, accurate clock synchronization is carried out, the metering modules at the branch side and the meter box side are checked for roll calling one by one, and if clock second-level deviation exists, the clock deviation metering modules are re-calibrated; secondly, carrying out a research on a synchronous sampling technology of characteristic signals based on HPLC, when the voltage approaches over zero, superposing the voltage characteristic signals to realize information representation, and starting synchronous sampling after each stage of metering module receives the voltage characteristic signals in a voltage period;
step 2.3: the method comprises the steps that a high-frequency data compression transmission technology based on a hybrid networking technology is utilized to realize high-efficiency data transmission to a management terminal and a main station and carry out analysis and calculation, firstly, the high-frequency data compression transmission based on the hybrid networking technology is researched, a character string mode is replaced by numbers, a Huffman binary tree is constructed by using the occurrence frequency of characters, data with more occurrence times are arranged on the upper layer of the tree, and data with less occurrence times are arranged on the lower layer of the tree; coding is carried out on a path from the root node to each data, lossless compression is realized, and efficient transmission of the data is further realized according to a physical topological relation; and secondly, realizing the data sampling of the electrical parameter curve of the same time section based on a synchronous sampling technology.
4. The model for the informatization evaluation and calibration of the electric energy meter based on the conservation of electric energy of the transformer area according to claim 1, wherein the step 3 specifically comprises the following steps:
step 3.1: the electric energy meter informatization evaluation calibration problem is accurately mapped by using a non-convex sparse multi-target optimization mathematical model construction method for the electric energy meter; based on the electric energy conservation principle of a transformer area, constructing a multi-objective optimization model which takes errors of an electric energy meter as variables and takes data priorality, system stability, energy conservation and the like as a plurality of optimization targets, and describing the data priorality targets by adopting a measurement function with non-convex and sparse mathematical characteristics; a measurement function with non-smooth mathematical characteristics is adopted to depict a system stability target; a measurement function with convex and smooth mathematical characteristics is adopted to depict an energy conservation target; finally, a non-convex sparse multi-target optimization model for accurately mapping the electric energy meter informatization evaluation calibration problem is constructed;
step 3.2: the original problem is reasonably simplified by using an effective relaxation method of a non-convex sparse multi-target optimization mathematical model of the electric energy meter; aiming at the non-convex, non-smooth and multi-target mathematical characteristics of the non-convex sparse multi-target optimization model, researching various convex methods of non-convex targets and corresponding convex approximation theory, and describing the error between the original problem solution and the relaxation problem solution; the method comprises the steps of describing the error between an original problem solution and a relaxation problem solution by utilizing a smoothing method of a non-smooth sparse target and a deep learning method of various smoothing model parameters; based on various existing scaling methods, an autonomous learning strategy for parameters among optimization targets is designed, and a new scaling method is developed;
step 3.3: the method comprises the steps that an efficient self-adaptive algorithm of a non-convex sparse multi-target optimization mathematical model of the electric energy meter is utilized to realize accurate identification of the out-of-tolerance electric energy meter; aiming at a special sparse structure of a model, a machine learning technology is combined, an efficient solving algorithm adaptive to a special sparse structure multi-target optimization problem is developed, the efficient solving algorithm is designed by using the calculation experience of the existing global optimization algorithm for reference, a local jump-out mechanism of a non-convex multi-target optimization problem global self-adaptive optimization algorithm is designed, and a better local optimal solution or global optimal solution is sought; the search direction and step length design of an approximation algorithm, a conjugate gradient algorithm and a confidence domain algorithm are utilized, so that the accuracy and the efficiency of the algorithm are improved; and further developing the theoretical analysis of algorithm convergence on the basis of the existing theoretical analysis results.
5. The electric energy meter informatization evaluation calibration model based on station area electric energy conservation according to claim 1, wherein the step 4 specifically comprises the following steps:
step 4.1: the method for estimating the line loss of the transformer area by utilizing dynamic and static data and a mathematical physical model to drive accurately describes the nonlinear function relationship among electric quantity, voltage, current, topology of the transformer area, line parameter dynamic and static data and line loss of the transformer area; calculating the voltage loss and the current loss of the current transformer area by using real-time dynamic data of voltage and current, static data of voltage impedance, current impedance and line parameters and a physical model of an electric power formula; by combining the calculated electric energy meter error data and utilizing a machine learning means, researching the nonlinear function relation among voltage loss, current loss, electric energy meter error, power supply and consumption difference, distribution room topological data and distribution room line loss, and realizing accurate estimation of the distribution room line loss;
step 4.2: constructing a multivariable coupling optimization model of the transformer area line loss and the electric energy meter operation error by utilizing a multivariable optimization model design method taking the transformer area line loss and the electric energy meter operation error as variables, combining a non-convex sparse optimization model of electric energy meter error estimation and a machine learning estimation model of the transformer area line loss;
step 4.3: solving the line loss of the transformer area and the running error of the electric energy meter by utilizing a separable alternative iteration algorithm and cooperative coupling; aiming at a multivariable optimization model, utilizing an augmented Lagrange function, a gradient function and a dual function thereof; decomposing the original problem into a plurality of optimization sub-problems by utilizing the separability of the objective function; constructing a distributed alternating direction multiplier method and a distributed block coordinate descent method, and solving optimization subproblems about line loss of a transformer area and errors of an electric energy meter in parallel; and constructing a self-adaptive shutdown criterion, accelerating the convergence speed of the algorithm, and analyzing the calculation complexity and the convergence of the algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211316897.XA CN115877312A (en) | 2022-10-26 | 2022-10-26 | Electric energy meter informatization evaluation calibration model based on station area electric energy conservation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211316897.XA CN115877312A (en) | 2022-10-26 | 2022-10-26 | Electric energy meter informatization evaluation calibration model based on station area electric energy conservation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115877312A true CN115877312A (en) | 2023-03-31 |
Family
ID=85758979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211316897.XA Pending CN115877312A (en) | 2022-10-26 | 2022-10-26 | Electric energy meter informatization evaluation calibration model based on station area electric energy conservation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115877312A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562655A (en) * | 2023-07-05 | 2023-08-08 | 北京理工大学 | Method, equipment and storage medium for designing flow flexible sparse structure |
| CN116718979A (en) * | 2023-08-08 | 2023-09-08 | 北京京仪北方仪器仪表有限公司 | Smart electric meter operation error measurement method and system |
-
2022
- 2022-10-26 CN CN202211316897.XA patent/CN115877312A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116562655A (en) * | 2023-07-05 | 2023-08-08 | 北京理工大学 | Method, equipment and storage medium for designing flow flexible sparse structure |
| CN116562655B (en) * | 2023-07-05 | 2023-09-15 | 北京理工大学 | Method, equipment and storage medium for designing flow flexible sparse structure |
| CN116718979A (en) * | 2023-08-08 | 2023-09-08 | 北京京仪北方仪器仪表有限公司 | Smart electric meter operation error measurement method and system |
| CN116718979B (en) * | 2023-08-08 | 2023-10-24 | 北京京仪北方仪器仪表有限公司 | Smart electric meter operation error measurement method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115877312A (en) | Electric energy meter informatization evaluation calibration model based on station area electric energy conservation | |
| Della Giustina et al. | Electrical distribution system state estimation: measurement issues and challenges | |
| CN113078630B (en) | Low-voltage distribution network topology identification method based on real-time measurement data | |
| KR101219545B1 (en) | Optimized parameter estimation method for power system | |
| CN107103184A (en) | A kind of high-voltage cable joint temperature predicting method | |
| CN103258103A (en) | Thevenin equivalent parameter identification method based on partial least squares regression | |
| CN110363334B (en) | Grid line loss prediction method of photovoltaic grid connection based on gray neural network model | |
| CN114091816A (en) | Power distribution network state estimation method based on gated graph neural network of data fusion | |
| CN111415068A (en) | Power distribution decision modeling method based on relevance of transformation measures and loss load index | |
| CN110543696A (en) | Method for enabling unmodeled small unit to participate in clearing and safety check of power market | |
| CN106372440B (en) | A kind of adaptive robust state estimation method of the power distribution network of parallel computation and device | |
| CN104537233B (en) | A kind of power distribution network puppet based on Density Estimator measures generation method | |
| Zhang et al. | STPGTN–A Multi-Branch Parameters Identification Method Considering Spatial Constraints and Transient Measurement Data. | |
| CN110707693A (en) | Ensemble Kalman filtering dynamic state estimation method based on AMI full-scale measuring point partition | |
| Pinto et al. | Reliability‐constrained robust expansion planning of active distribution networks | |
| CN112446175B (en) | Distributed power supply site selection optimization method based on power distribution network probability load flow calculation | |
| Ajoudani et al. | Improving state estimation accuracy in active distribution networks by coordinating real‐time and pseudo‐measurements considering load uncertainty | |
| Hayes et al. | Short-term operational planning and state estimation in power distribution networks | |
| CN110752622A (en) | Power distribution network affine state estimation method | |
| EP4254705A1 (en) | Computer-implemented method or assessment of a power distribution grid | |
| CN115600722A (en) | Long-term prediction method for electric carbon factor of enterprise | |
| Nusrat | Development of novel electrical power distribution system state estimation and meter placement algorithms suitable for parallel processing | |
| CN114742258B (en) | Migratable residential household power load scene prediction method | |
| CN117277446B (en) | Multi-target power distribution network planning method and system | |
| CN115036923B (en) | Electricity-gas comprehensive energy system state estimation method considering multi-energy flow time sequence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |