Electric energy meter state evaluation method and system based on optimized kernel support vector regression

Abstract

The application discloses a kind of based on optimization kernel support vector regression's electric energy meter state evaluation method and system, method includes: the operating state data set of electric energy meter is collected;Mapping kernel function is constructed, and optimization kernel function is obtained based on mapping kernel function and modification kernel;Optimization kernel support vector regression model is established;Global optimization is carried out to the parameter in optimization kernel support vector regression model using particle swarm algorithm;With the root mean square error of training set as fitness function, iteratively update particle position and speed until meeting convergence condition, obtain the best parameter combination;The best parameter combination is substituted into optimization kernel support vector regression model, and the model is trained and the prediction performance is evaluated;Measurement error prediction is carried out using the optimization kernel support vector regression model trained, and the influence degree of operating time, temperature stress and humidity stress on the operating state of electric energy meter is respectively evaluated according to the size of kernel coefficient.The application has the advantages of high evaluation accuracy.

Description

Technical Field

[0001] This invention mainly relates to the field of smart energy meter status assessment and reliability analysis under typical environments, specifically to an energy meter status assessment method and system based on optimized kernel support vector regression. Background Technology

[0002] Given the vast geographical area and diverse climates, the impact of different regional natural environments on the operational status of smart meters varies significantly. Accurate metering and reliable operation of smart meters under typical natural environments are current research hotspots. Typical natural environments include extreme climates such as high altitude, extreme cold, and high salt spray. Current research largely focuses on studying the operational characteristics and failure mechanisms of smart meters in environmental impact test chambers, neglecting their operation and failure analysis in real-world field environments. Tests in environmental impact test chambers cannot accurately and completely reproduce the operational characteristics and failure mechanisms of smart meters under multiple environmental stresses in typical natural environments. Therefore, evaluating the operational status of smart meters under typical natural environments is crucial for guiding equipment selection and maintenance, effectively saving production costs, improving energy conversion efficiency, and ultimately promoting the development of smart grids.

[0003] Currently, the methods for assessing the condition of instruments and equipment are mainly divided into two categories: direct assessment and data-driven condition assessment methods. Direct assessment methods utilize reliability analysis manuals and degradation data from various components of a smart meter to directly evaluate its operational status. Commonly used direct assessment methods include expert systems and decision trees. These methods are simple to operate and easy to implement, but require prior knowledge of the degradation parameters of the instrument and equipment, referring to the reliability analysis manual. Furthermore, because they require fault or degradation data from multiple components of the equipment, they increase the actual data acquisition cost, making them difficult to apply to the operational status analysis of high-level metering equipment like smart meters. In contrast, data-driven analysis methods that only require overall equipment performance data have greater applicability.

[0004] Data-driven approaches utilize accumulated operational status data and employ intelligent algorithms such as machine learning and deep learning to assess equipment operational status. These intelligent algorithms include hierarchical Bayesian, Gaussian process regression, neural networks, and support vector regression. Hierarchical Bayesian and Gaussian process regression analyses can fully consider equipment uncertainties and make assumptions about the failure mechanism of the equipment through appropriate prior distributions, thereby reasonably evaluating the input features of the model. However, in practical applications, the degradation trends of equipment vary greatly under different scenarios, and the inherent prior distributions are difficult to meet the needs of equipment status assessment in typical environments. Neural networks and support vector regression do not require prior assumptions about the failure mechanism, and are therefore suitable for equipment status assessment problems involving multi-feature fusion. However, the data consumption during the training process of neural network models is large, making them unsuitable for small sample data analysis scenarios. Support vector regression has the advantages of simple structure and strong small sample learning ability, but traditional single empirical kernels are difficult to effectively integrate multiple stress features for accurate assessment.

[0005] In conclusion, accurate assessment of the operational status of smart meters under typical natural environments is imperative. Data-driven assessment methods are currently the mainstream research direction, with different intelligent analysis algorithms each having their own advantages and disadvantages. Further exploration is needed to fully utilize multiple environmental stress characteristics to provide accurate and reliable operational status assessment results under conditions of small sample smart meter operational data. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention provides a method and system for assessing the state of electricity meters based on optimized kernel support vector regression, which has high assessment accuracy.

[0007] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows: A method for assessing the state of electricity meters based on optimized kernel support vector regression, comprising the following steps: S1. Collect the operating status dataset of the energy meter under test in a typical natural environment; the operating status dataset includes operating time, temperature stress, humidity stress, and smart energy meter measurement error; S2. Assign a polynomial kernel function to the running time, and an RBF kernel function to the temperature stress and humidity stress, and construct a mapping kernel function by linear combination; then construct a modified kernel using the measurement error; obtain the optimized kernel function based on the mapping kernel function and the modified kernel. S3. Based on the optimized kernel function, an optimized kernel support vector regression model is established. By introducing relaxation factors and constraints, the optimization problem of the optimized kernel support vector regression model is transformed into a Lagrange dual form, and the optimal decision function is derived for predicting measurement errors. S4. Use the particle swarm optimization algorithm to globally optimize the parameters in the kernel support vector regression model; use the root mean square error of the training set as the fitness function, iteratively update the particle position and velocity until the convergence condition is met, and obtain the optimal parameter combination. S5. Substitute the optimal parameter combination into the optimized kernel support vector regression model, train the model, and use cross-validation to evaluate the predictive performance of the model, and record the kernel coefficients of each kernel function; S6. Use the trained optimized kernel support vector regression model to predict measurement errors, and evaluate the impact of running time, temperature stress, and humidity stress on the operating status of the electricity meter according to the kernel coefficients, and output the evaluation results.

[0008] Preferably, the optimized kernel function in step S2 Specifically:

[0009] in For mapping kernel functions; For modifying the kernel.

[0010] Preferably, the mapping kernel function for:

[0011] in, Represents the polynomial kernel function. Represents the RBF kernel function. These represent the kernel coefficients corresponding to operating time, temperature stress, and humidity stress, respectively. Polynomial kernel function The expression is as follows:

[0012] in, d Let be the degree of the polynomial kernel function; For the running state dataset Any two data points in the middle; RBF kernel function The expression is as follows:

[0013] in, The radius of the RBF core.

[0014] Preferably, in step S3, the optimization problem of optimizing the kernel support vector regression model is:

[0015] The constraints are:

[0016] in, For decision-making surface weights, For bias terms, For the feature space, The penalty coefficient is... Insensitive loss; For the input sample, For the first The true output value of each training sample; Corresponding to the The relaxed variables of each training sample; Corresponding to the The underslack variables of each sample; This represents the total number of training samples.

[0017] Preferably, the optimal decision function is:

[0018] in, This is the predicted value for measurement error; For the first Lagrange multipliers of each training sample For the first Lagrange multipliers of each training sample; To optimize the kernel function; This is a bias term.

[0019] Preferably, in step S4, optimizing the parameters in the kernel support vector regression model includes the degree of the multinomial kernel function. d RBF nuclear radius nuclear modification factors λ、 and the kernel coefficients corresponding to operating time, temperature stress, and humidity stress. .

[0020] Preferably, in step S4, the parameters of the particle swarm algorithm are set as follows: population size is 100, number of iterations is 200, inertia weight is 0.8, acceleration coefficient is 1.5, maximum speed is 10, and minimum speed is -10.

[0021] Preferably, in step S4, the fitness function for:

[0022] in, This represents the actual output data of the training set. For model prediction data; The total number of training samples; RMSE The root mean square error of the training set.

[0023] Preferably, in step S5, the cross-validation is five-fold cross-validation; the evaluation metric for prediction performance includes root mean square error. RMSE and coefficient of determination .

[0024] The present invention also discloses an energy meter status assessment system based on optimized kernel support vector regression, comprising an interconnected memory and a processor, wherein the memory stores a computer program, and the computer program executes the steps of the method described above when run by the processor.

[0025] Compared with the prior art, the advantages of the present invention are as follows: To address the problem of insufficient utilization of feature information and low prediction accuracy in the evaluation of smart meter operation status under typical natural environments, this invention combines operational status data of smart meters at high-altitude test bases. Based on the data characteristics of different input stresses, it selects an appropriate mapping kernel function to fuse time stress and environmental stress in a high-dimensional feature space. During training, the kernel function is modified using output measurement error information to improve its prediction accuracy. Finally, while predicting the measurement error of the smart meter, the kernel coefficient of the stress is used to analyze its actual impact on the smart meter's operational status. Based on the evaluation results of this invention, theoretical guidance can be provided for the selection and upgrading of smart meters in typical regions. Furthermore, this invention is applicable not only to smart meters but also to the evaluation of the operational status of other metering equipment under different typical environments, demonstrating strong versatility.

[0026] The method of this invention includes performing kernel function matching on the operating time of a smart energy meter under typical environmental stress and the environmental stress, and assigning a polynomial kernel function to the time stress. K p RBF kernel function for stress distribution based on temperature and humidity K r An initial mapping kernel function is constructed by linearly combining three kernel functions to achieve information fusion under multiple inputs. Then, based on the initial mapping kernel function, the kernel function is reconstructed and optimized using the output information, i.e., the measurement error data of the smart energy meter. K This invention creates an optimized kernel support vector regression model, using a particle swarm optimization algorithm with strong global search capabilities to optimize model parameters. This enables the modeling of smart meter degradation analysis and utilizes different kernel coefficients to assess the impact of different input information on the smart meter's operating status. The invention can accurately predict the degradation trend of smart meters under typical environments and provides guidance for the design and maintenance of smart meters based on the weighting of different stress factors.

[0027] This invention overcomes the limitation of traditional support vector regression's single kernel function in simultaneously achieving both learning and generalization capabilities for multi-feature fusion. Based on the characteristics of the input features, different kernel functions are used for mapping to achieve effective fusion of multiple features in a high-dimensional space. At the same time, during model training, the kernel function is modified using the output information of the training samples to further improve the accuracy of the evaluation results. Attached Figure Description

[0028] Figure 1 This is a flowchart of the electricity meter status assessment method according to an embodiment of the present invention.

[0029] Figure 2 This is a schematic diagram illustrating the process of optimizing model parameters using the particle swarm optimization algorithm in an embodiment of the present invention. Detailed Implementation

[0030] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0031] like Figure 1 As shown in the figure, the energy meter status assessment method based on optimized kernel support vector regression provided by this invention includes the following steps: Step S1. Collect the data set of the operating status of the energy meter under test in a typical natural environment; Smart meters were deployed to a typical environmental test site for field operation testing, allowing the smart meters under test to operate in real-time under typical natural conditions. Power consumption was generated using a standard source at the test site, and the standard meter measured the measurement error of the smart meter under test. Temperature and humidity sensors were used to record the temperature and humidity stress values ​​of the smart meters. Monthly measurements were collected, and the field test data was uploaded to a data acquisition system to store and accumulate a dataset of the smart meters' operational status under typical natural conditions. ,in, The running time is defined as follows: the initial value is 0 when the device is first put into testing, and the difference between the initial value and the subsequent time is taken as the running time of the smart meter. In months; For temperature stress, For humidity stress, This refers to the measurement error data of smart energy meters; For the original dataset To process, in order to Input variables , Its dimension is 3, and the measurement error is the output variable. , Its dimension is 1; To reduce the impact of differences in data magnitude on model results, the Z-score method is used to perform normalization processing on the input data, that is... The numerical range of the input variable is transformed to [0,1]; thus, the normalized input variable... Output variables ; Step S2. Assign a polynomial kernel function to the running time, and assign RBF kernel functions to the temperature stress and humidity stress. Construct a mapping kernel function by linear combination. Then construct a modified kernel using the measurement error. Obtain the optimized kernel function based on the mapping kernel function and the modified kernel. At the input level, to fully consider the differences in stress characteristics, different kernel functions are selected to map the input features. Specifically: due to runtime... It exhibits a linear increasing trend, therefore a polynomial kernel function is used to characterize its linear variation trend; temperature stress Humidity stress It exhibits a nonlinear variation trend with an approximate annual period, therefore its local irregularity characteristics are learned using the RBF kernel function; For dataset Any two data points in the training sample Mapping kernel function It can be represented as:

[0032] in, Represents the polynomial kernel function. Represents the RBF kernel function. These represent the kernel coefficients corresponding to running time, temperature stress, and humidity stress, respectively, reflecting the degree of influence of different input variables on the final output result.

[0033] Polynomial kernel function The expression is as follows:

[0034] in, d Let be the degree of the polynomial kernel function.

[0035] RBF kernel function The expression is as follows:

[0036] in, The radius of the RBF core.

[0037] After establishing the mapping relationship between input features, to further improve the evaluation accuracy of the support vector regression model, output information is used to enhance the mapping performance of the kernel function during training. This is done by mapping the kernel function... Based on this, a modified kernel is introduced. ,have:

[0038] in, As a modifier, it determines the degree to which the output information affects the kernel function.

[0039] Based on mapping kernel function With modified kernel Optimized kernel function Represented as: .

[0040] Step S3. Based on the optimized kernel function, establish an optimized kernel support vector regression model, introduce relaxation factors and constraints, transform the optimization problem of the optimized kernel support vector regression model into a Lagrange dual form, and derive the optimal decision function for predicting measurement errors; Determining the optimal kernel function Then, a relaxation factor was introduced. :

[0041] Combined with optimized kernel function The optimization problem of the model can be expressed as:

[0042] The constraints are satisfied:

[0043] in, For decision-making surface weights, For bias terms, For the feature space, The penalty coefficient is... Insensitive loss; For input samples; For the first The true output value (measurement error) of each sample; where Corresponding to the The upper-slack variables of each sample; Corresponding to the The underslack variables of each sample; This represents the total number of training samples.

[0044] According to the Neumann divergence, Two data points in the test sample Optimized kernel function The solution is as follows:

[0045] in: ; Finally, by introducing the Lagrange dual function, we derive the optimal decision function for optimizing kernel support vector regression in the test samples as follows:

[0046] in, This is the predicted value for measurement error; For Lagrange multipliers, among which For the first Lagrange multipliers of each training sample For the first Lagrange multipliers of each training sample; To optimize the kernel function; This is a bias term.

[0047] Step S4. Use the particle swarm optimization algorithm to globally optimize the parameters in the kernel support vector regression model; use the root mean square error of the training set as the fitness function to iteratively update the particle position and velocity until the convergence condition is met, and obtain the optimal parameter combination. like Figure 2 As shown, the particle swarm optimization algorithm is used to optimize the model parameters in kernel support vector regression, including the degree of the polynomial kernel function. d RBF nuclear radius nuclear modification factors λ、 and the kernel coefficients corresponding to operating time, temperature stress, and humidity stress. ; The parameters of the particle swarm optimization algorithm are set as follows: population size is 100, number of iterations is 200, inertia weight is 0.8, two acceleration coefficients are set to 1.5, maximum speed is 10, and minimum speed is -10. Regarding the determination of fitness in the particle swarm optimization algorithm, this invention uses the root mean square error (RMSE) of the optimized kernel support vector regression as the fitness function of the algorithm. The expression is:

[0048] in, Indicates the first The actual output data of each training sample Indicates the first Predicted data for each training sample; The index for the training samples; Based on the determined fitness function, the iteration ends when the difference before and after the root mean square error meets a given threshold, or when the number of iterations reaches a specified value. In this invention, the threshold is set to 0.0000001. The optimal parameter values ​​obtained by the algorithm are used as the final parameters of the optimized kernel support vector regression model.

[0049] Step S5. Substitute the optimal parameter combination into the optimized kernel support vector regression model, train the model, and use cross-validation to evaluate the predictive performance of the model, and record the kernel coefficients of each kernel function; The simulation experiments were conducted using MATLAB, with a support vector regression model built using LibSVM. The experimental dataset was divided into training and test sets in a 4:1 ratio. The model was trained using the training set data, employing five-fold cross-validation to fully utilize the training data.

[0050] Based on the optimized kernel function and optimal parameters determined in steps S2 to S4, an optimized kernel support vector regression model is constructed, and the kernel coefficients corresponding to different kernel functions are recorded. .

[0051] Step S6. Use the trained optimized kernel support vector regression model to predict measurement errors, and evaluate the impact of running time, temperature stress and humidity stress on the operating status of the electricity meter according to the size of the kernel coefficients, and output the evaluation results.

[0052] For the trained model, the root mean square error (RMSE) and coefficient of determination are used as the basis for the evaluation. The final performance of the optimized kernel support vector regression was tested on the new test set data as an evaluation metric.

[0053] In this embodiment, the optimization results of the kernel coefficients of the three kernel functions and the model prediction performance are shown in Table 1. Table 1. Model kernel coefficients and prediction performance

[0054] Table 1 shows that the optimized kernel support vector regression model has accurate predictive performance, with an RMSE of only 0.018 and an R² of 0.984. This indicates that the model can accurately predict the changing trend of smart meter measurement errors under this typical environment. Furthermore, while accurately predicting error changes, it also reflects the different forces exerted on the model by the input information. The optimal values ​​of the three kernel coefficients show that the kernel coefficient corresponding to temperature stress has the highest value of 0.92, while the optimal value corresponding to time stress is only 0.18. This indicates that in this typical environmental test base, temperature characteristics have the highest influence on the measurement error of smart meters, meaning they have the strongest determining effect on the stability of their operating state. Based on the experimental results, the design and modification of this type of smart meter should strengthen its temperature adaptability to improve its operational reliability under this typical environment.

[0055] In summary, addressing the issues of insufficient utilization of feature information and low prediction accuracy in the evaluation of smart meter operating status under typical natural environments, this invention combines operating status data of smart meters at high-altitude test bases. Based on the data characteristics of different input stresses, it selects an appropriate mapping kernel function to fuse time stress and environmental stress in a high-dimensional feature space. During training, the kernel function is modified using output measurement error information to improve its prediction accuracy. Finally, while predicting the measurement error of the smart meter, the kernel coefficient of the stress is used to analyze its actual impact on the operating status of the smart meter. Based on the evaluation results of this invention, theoretical guidance can be provided for the selection and upgrading of smart meters in typical regions. Furthermore, this invention is applicable not only to smart meters but also to the evaluation of the operating status of other metering equipment under different typical environments, demonstrating strong versatility.

[0056] The method of this invention includes performing kernel function matching on the operating time of a smart energy meter under typical environmental stress and the environmental stress, and assigning a polynomial kernel function to the time stress. K p RBF kernel function for stress distribution based on temperature and humidity K r An initial mapping kernel function is constructed by linearly combining three kernel functions to achieve information fusion under multiple inputs. Then, based on the initial mapping kernel function, the kernel function is reconstructed and optimized using the output information, i.e., the measurement error data of the smart energy meter. K This invention creates an optimized kernel support vector regression model, using a particle swarm optimization algorithm with strong global search capabilities to optimize model parameters. This enables the modeling of smart meter degradation analysis and utilizes different kernel coefficients to assess the impact of different input information on the smart meter's operating status. The invention can accurately predict the degradation trend of smart meters under typical environments and provides guidance for the design and maintenance of smart meters based on the weighting of different stress factors.

[0057] This invention overcomes the limitation of traditional support vector regression's single kernel function in simultaneously achieving both learning and generalization capabilities for multi-feature fusion. Based on the characteristics of the input features, different kernel functions are used for mapping to achieve effective fusion of multiple features in a high-dimensional space. At the same time, during model training, the kernel function is modified using the output information of the training samples to further improve the accuracy of the evaluation results.

[0058] This invention also discloses a power meter condition assessment system based on optimized kernel support vector regression, comprising an interconnected memory and a processor. The memory stores a computer program, which, when run by the processor, executes the steps of the method described above. The power meter condition assessment system of this invention, corresponding to the aforementioned assessment method, also possesses the advantages described above.

[0059] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium includes: any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0060] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for assessing the state of an energy meter based on optimized kernel support vector regression, characterized in that, Including the following steps: S1. Collect the operating status dataset of the energy meter under test in a typical natural environment; the operating status dataset includes operating time, temperature stress, humidity stress, and smart energy meter measurement error; S2. Assign a polynomial kernel function to the running time, and an RBF kernel function to the temperature stress and humidity stress, and construct a mapping kernel function by linear combination; then construct a modified kernel using the measurement error; obtain the optimized kernel function based on the mapping kernel function and the modified kernel. S3. Based on the optimized kernel function, an optimized kernel support vector regression model is established. By introducing relaxation factors and constraints, the optimization problem of the optimized kernel support vector regression model is transformed into a Lagrange dual form, and the optimal decision function is derived for predicting measurement errors. S4. Use the particle swarm optimization algorithm to globally optimize the parameters in the kernel support vector regression model; use the root mean square error of the training set as the fitness function, iteratively update the particle position and velocity until the convergence condition is met, and obtain the optimal parameter combination. S5. Substitute the optimal parameter combination into the optimized kernel support vector regression model, train the model, and use cross-validation to evaluate the predictive performance of the model, and record the kernel coefficients of each kernel function; S6. Use the trained optimized kernel support vector regression model to predict measurement errors, and evaluate the impact of running time, temperature stress and humidity stress on the operating status of the electricity meter according to the size of the kernel coefficients, and output the evaluation results. Optimizing the kernel function in step S2 Specifically: in For mapping kernel functions; For modifying the nucleus; Mapping kernel function for: in, Represents the polynomial kernel function. Represents the RBF kernel function. These represent the kernel coefficients corresponding to operating time, temperature stress, and humidity stress, respectively. Polynomial kernel function The expression is as follows: in, d Let be the degree of the polynomial kernel function; For the running state dataset Any two data points in the middle; RBF kernel function The expression is as follows: in, The radius of the RBF core; In step S4, optimizing the parameters in the kernel support vector regression model includes the degree of the multinomial kernel function. d RBF nuclear radius nuclear modification factors λ、 and the kernel coefficients corresponding to operating time, temperature stress, and humidity stress. .

2. The method for assessing the state of an energy meter based on optimized kernel support vector regression according to claim 1, characterized in that, In step S3, the optimization problem for the kernel support vector regression model is: The constraints are: in, For decision-making surface weights, For bias terms, For the feature space, The penalty coefficient is... Insensitive loss; For the input sample, For the first The true output value of each training sample; Corresponding to the The relaxed variables of each training sample; Corresponding to the The underslack variables of each sample; This represents the total number of training samples.

3. The method for assessing the state of an energy meter based on optimized kernel support vector regression according to claim 2, characterized in that, The optimal decision function is: in, This is the predicted value for measurement error; For the first Lagrange multipliers of each training sample For the first Lagrange multipliers of each training sample; To optimize the kernel function; This is a bias term.

4. The method for assessing the state of an energy meter based on optimized kernel support vector regression according to claim 1, 2, or 3, characterized in that, In step S4, the parameters of the particle swarm algorithm are set as follows: population size is 100, number of iterations is 200, inertia weight is 0.8, acceleration coefficient is 1.5, maximum speed is 10, and minimum speed is -10.

5. The method for assessing the state of an energy meter based on optimized kernel support vector regression according to claim 1, 2, or 3, characterized in that, In step S4, the fitness function for: in, This represents the actual output data of the training set. For model prediction data; The total number of training samples; RMSE The root mean square error of the training set.

6. The method for assessing the state of an energy meter based on optimized kernel support vector regression according to claim 1, 2, or 3, characterized in that, In step S5, the cross-validation is five-fold cross-validation; the evaluation metrics for prediction performance include root mean square error. RMSE and coefficient of determination .

7. A power meter condition assessment system based on optimized kernel support vector regression, comprising an interconnected memory and a processor, wherein the memory stores a computer program, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1-6.

Images