Power load anomaly detection method and system based on time sequence correlation deviation

Abstract

The invention discloses a power load anomaly detection method and system based on time sequence correlation deviation, and relates to the technical field of power data analysis, and the method comprises the steps: collecting time sequence data used for power load anomaly detection; performing feature reconstruction on the time sequence data through a self-attention neural network, and obtaining corresponding reconstructed sequence data; obtaining an incidence matrix in the self-attention neural network feature reconstruction process, and calculating a time sequence correlation deviation of the time sequence data based on the incidence matrix; based on the time sequence data and the corresponding reconstruction sequence data, mutual reconstruction errors are obtained through calculation; the time sequence correlation deviation and the reconstruction error are utilized to calculate an anomaly score of the time sequence data, and anomaly judgment is carried out based on a set anomaly threshold value; the method does not need to depend on labeled data, can adaptively learn a long-term dependence structure of load data, accurately recognizes load abnormal points, and provides a more efficient solution for safety monitoring of a smart power grid.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power data analysis, more particularly, it relates to a power load anomaly detection method and system based on time series correlation deviation. BACKGROUND

[0002] Power load anomaly detection is a key link to ensure the stable operation of the power system, aiming to identify load fluctuation anomalies, equipment failures, control strategy failures, etc., to prevent power grid operation risks. Power load data is usually recorded in the form of time series, with obvious trend, periodicity and short-term fluctuation characteristics. Under normal circumstances, load data shows regular changes, such as seasonal fluctuations, day-night load differences, and load pattern differences between weekdays and holidays. However, abnormal situations can disrupt these regularities, resulting in sudden abnormal points. For example, short-term mutation anomalies refer to sudden increases or decreases in load within a very short period of time, trend drift anomalies show that the load continuously deviates from the normal trajectory, periodic anomalies change the fluctuation pattern due to external interference, and random noise anomalies are caused by external environment or data collection errors. These anomalies not only affect the accuracy of power dispatching, but also pose a threat to equipment safety.

[0003] Power load anomaly detection has important practical significance and application value in the smart grid environment. First, accurate identification of abnormal load can help operations personnel quickly locate the source of the fault, prevent equipment damage or fault propagation, and thus ensure the safe and stable operation of the power grid. Second, anomaly detection can provide early warning of potential risks before faults occur, providing decision support for power dispatching and optimization, and improving power supply reliability. In addition, combined with big data and artificial intelligence technology, an intelligent anomaly detection system can reduce manual intervention, achieve automated monitoring and intelligent operation and maintenance of power system equipment, improve detection efficiency and reduce operation and maintenance costs.

[0004] Traditional anomaly detection methods mainly rely on fixed rules or statistical analysis, which are difficult to effectively model complex time series patterns, especially in capturing long-term dependencies. With the expansion of the power grid scale and the improvement of the level of intelligence, load anomaly detection faces more severe challenges, including multi-scale time series dependencies, multi-dimensional feature interactions, and data uncertainty and noise interference, making it difficult for traditional methods to meet the high-precision detection requirements. Therefore, time series modeling methods based on deep learning have gradually become a research hotspot. Among them, Transformer and attention mechanism have advantages in capturing long-term dependencies and have achieved remarkable results in natural language processing and other fields, providing new solutions for complex time series anomaly detection.

[0005] In recent years, reconstruction-based unsupervised anomaly detection methods have shown good results in many fields. These methods assume that normal data can be accurately reconstructed, while the reconstruction error of abnormal data is larger, thereby achieving anomaly point identification. However, the existing methods still have the following two main problems: (1) Dependence on normal samples: Many methods filter normal samples for training through heuristic rules, but this may introduce additional noise and affect detection accuracy; (2) Limited time series modeling capability: Traditional methods mainly rely on recurrent neural networks (RNN) and their variants to model time series, but these methods have limitations in learning long-term dependencies and periodic features, making it difficult to adapt to long-span power load anomaly detection requirements. SUMMARY

[0006] The purpose of the present application is to provide a power load anomaly detection method and system based on time series correlation deviation, which does not rely on labeled data, can adaptively learn the long-term dependency structure of load data, and accurately identify load anomaly points, providing a more efficient solution for the safety monitoring of smart grids.

[0007] The above technical purpose of the present application is achieved by the following technical scheme: In a first aspect, the present application provides a power load anomaly detection method based on time series correlation deviation, comprising the following specific steps: S1, collecting time series data for power load anomaly detection and preprocessing the time series data; S2, reconstructing the features of the preprocessed time series data through a self-attention neural network and obtaining the corresponding reconstructed sequence data; S3, obtaining the correlation matrix in the feature reconstruction process of the self-attention neural network and calculating the time series correlation deviation of the time series data based on the correlation matrix; S4, calculating the reconstruction error between the time series data and the corresponding reconstructed sequence data; S5, calculating the anomaly score of the time series data using the time series correlation deviation and the reconstruction error, and performing anomaly judgment based on a set anomaly threshold.

[0008] On the basis of the above technical scheme, the present application can be further improved as follows. Further, the preprocessing includes normalization, specifically: ; wherein, is the original time series data, wherein, is the number of samples of the time series data, denotes the number of time steps of each time series; denotes the time series data after normalization, and denote the mean and standard deviation of each time series, respectively.

[0009] Further, the reconstructed sequence data is specifically: ; In the formula, is the reconstructed sequence data, , , denotes the dimension of the output feature, denotes the dimension of the input feature; are the weight matrix and bias vector of the decoder in the self-attention neural network, respectively, denotes the output of the feedforward network in the self-attention neural network.

[0010] Further, the correlation matrix represents the long and short term dependency between time steps, and is specifically: ; In the formula, is the correlation matrix, is the attention score matrix of the attention subnetwork in the self-attention neural network, is the number of attention subnetworks in the self-attention neural network.

[0011] Further, the time series correlation bias is specifically: ; In the formula, denotes the time series correlation bias of the sample in the time series data, denotes the JS divergence function, is the average of the correlation matrix of all normal samples, is the correlation matrix corresponding to the sample .

[0012] Further, the self-attention neural network is obtained by the following method: input the training time series samples into the neural network for processing, and calculate the time series correlation bias loss and reconstruction error loss of the training time series samples; based on the time series correlation bias loss and the reconstruction error loss, calculate the training target function of the neural network, when the training target function does not exceed the loss threshold, determine the neural network whose training target function does not exceed the loss threshold as the self-attention neural network; the training target function is specifically: wherein: ; ; In the formula, is the value of the training target function, represents the reconstruction error loss, is the time correlation deviation loss, is the weight parameter, is the time correlation deviation of the sample in the time series data, is the reconstruction error of the sample in the time series data, represents the total number of samples in the training time series sample, represents the reconstruction sequence data of the sample in the time series data.

[0013] Further, the above-mentioned anomaly score is specifically: ; In the formula, represents the anomaly score of the sample in the time series data, represents the reconstruction error of the sample in the time series data, represents the reconstruction sequence data of the sample in the time series data, is the time correlation deviation of the sample in the time series data.

[0014] In a second aspect, the present application provides a power load anomaly detection system based on time correlation deviation, which is applied to the power load anomaly detection method based on time correlation deviation in any one of the first aspect, comprising: a time series data acquisition module for collecting time series data for power load anomaly detection and preprocessing the time series data; a time series data reconstruction module for reconstructing the features of the preprocessed time series data through a self-attention neural network and obtaining corresponding reconstruction sequence data; a correlation deviation calculation module for obtaining a correlation matrix in the feature reconstruction process of the self-attention neural network and calculating the time correlation deviation of the time series data based on the correlation matrix; a reconstruction error calculation module for calculating the reconstruction error between the time series data and the corresponding reconstruction sequence data; an anomaly score judgment module for calculating the anomaly score of the time series data using the time correlation deviation and the reconstruction error, and performing anomaly judgment based on a set anomaly threshold.

[0015] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the method of any one of the first aspect when executing the computer program.

[0016] In a fourth aspect, the present application provides a non-transitory computer readable storage medium, which stores computer instructions, and the computer instructions make a computer execute the method of any one of the first aspect.

[0017] Compared with the prior art, the present application has at least the following beneficial effects: In the present application, firstly, the long-distance dependence relationship in the time series data is captured by using the self-attention mechanism, and the global pattern of load change is learned, the multi-level time series information is extracted by paying attention to multiple different positions in the input sequence; then, by constructing a time series correlation matrix, the time series correlation deviation between each time step is calculated, so as to quantify the change of the dependence relationship between the time steps; finally, the reconstruction error and the time series correlation deviation are used to calculate the abnormal score of the sample; the reconstruction error and the time series correlation deviation are fused, the stability of the abnormal detection is improved, compared with the traditional method which often depends on a single feature, the information of the two is comprehensively used, so that the detection result is more stable and accurate, and is suitable for complex power load change.

[0018] In the present application, the self-attention neural network adopts a multi-head self-attention mechanism, which can capture long-time span load change patterns, and make up for the defect that RNN and LSTM are difficult to model long-range dependence due to gradient vanishing; in contrast, the self-attention mechanism can pay attention to multiple time steps at the same time, improve the perception ability of complex patterns, and more accurately depict the change trend of the load data; at the same time, the neural network adopts an unsupervised training strategy, reduces the dependence on labeled data, and the power load abnormal sample is scarce and difficult to label, and the traditional supervised learning method is limited by insufficient samples and is difficult to generalize; the method jointly optimizes the reconstruction error and the time series correlation deviation, accurately learns the normal mode, and identifies abnormal load points based on the abnormal score mechanism, has good generalization ability, and is suitable for power data of different load modes and time scales.

[0019] In the present application, the time series correlation deviation is introduced as an important measurement index for anomaly detection, which makes up for the shortcomings of the traditional reconstruction error method, especially for the data points with small reconstruction error but abnormal dependence relationship; the method calculates the time series correlation matrix and uses the JS divergence to measure the deviation, accurately depicts the correlation change between the time steps, and improves the comprehensiveness and accuracy of the anomaly detection. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings, which are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings: Figure 1 A method flow chart of the detection method in the embodiments of the application; Figure 2 A flow chart of the detection method in the embodiments of the application; Figure 3 A model structure schematic diagram of the power load anomaly detection method in the embodiments of the application; Figure 4 A connection schematic diagram of the detection system in the embodiments of the application. DETAILED DESCRIPTION

[0021] To make the objects, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described below in connection with the drawings of the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, rather than all the embodiments. The components of the embodiments of the application described and shown in the drawings can be arranged and designed in various different configurations.

[0022] Therefore, the following detailed description of the embodiments of the application provided in the drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the application without creative labor are within the scope of protection of the application.

[0023] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0024] In the description of the embodiments of the application, "a plurality of" represents at least 2.

[0025] Embodiment 1: In order to solve the problems that the traditional network training relies on labeled data, and many methods train by screening normal samples through heuristic rules, which may introduce additional noise and affect detection accuracy, and the time series modeling capability is limited, which is difficult to adapt to the demand of long time span power load anomaly detection, etc., the embodiment provides a power load anomaly detection method based on time series correlation bias, as shown in Figure 1 and Figure 2 The embodiment includes the following specific steps: S1, collecting time series data for power load anomaly detection, and preprocessing the time series data.

[0026] The preprocessing includes normalization, specifically: ; wherein, is the original time series data, wherein, is the number of samples of the time series data, represents the number of time steps of each time series; represents the time series data after normalization processing, and respectively represent the mean and standard deviation of each time series.

[0027] S2, the pre-processed time series data is reconstructed by a self-attention neural network, and the corresponding reconstructed sequence data is obtained.

[0028] Optionally, the reconstructed sequence data is specifically: ; wherein, is the reconstructed sequence data, , , represents the dimension of the output feature, represents the dimension of the input feature; and respectively are the weight matrix and the bias vector of the decoder in the self-attention neural network, represents the output of the feedforward network in the self-attention neural network.

[0029] Specifically, when the feature reconstruction is performed by the self-attention neural network, the following steps are mainly included: 1) Linear embedding of the input time series data by linear mapping of the self-attention neural network.

[0030] The normalized time series data is embedded by the linear projection layer, which can be expressed as: ; wherein, is the data representation obtained after embedding, is a trainable parameter, is a bias term, wherein is the hidden layer dimension of the self-attention neural network.

[0031] 2) An encoder based on a multi-head self-attention network is constructed, which is composed of L layers of stacking, each layer containing a multi-head self-attention network and a feedforward network, as shown in Figure 3 wherein: The multi-head self-attention network comprises H sub-networks (referred to as "attention heads"); The multi-head self-attention network is represented as: wherein, represents the multi-head self-attention network, represents the output of the i-th layer; each attention sub-network comprises three functions: ; In the above formula, is the output of the i-th layer, , , , are three weight matrices respectively corresponding to the query function, the key function and the value function of the attention mechanism; each attention sub-network calculates an attention score as follows: ; In the formula, represents the attention score matrix of the attention sub-network , is a normalization function, is a scaling coefficient; the output of each attention sub-network is obtained by weighted summation of the value function using the attention score as follows: ; then the outputs of the multiple attention sub-networks are spliced to obtain the output of the multi-head self-attention network as follows: .

[0032] Further, the above-described feedforward network is composed of a fully connected layer and adds a residual connection and a layer normalization operation: ; In the formula, and are parameters of the feedforward network, is an activation function, represents a layer normalization operation.

[0033] 3) decoding by the decoder to obtain the final reconstructed sequence data, which adopts a fully connected layer for decoding to reconstruct the input time series.

[0034] S3, obtaining the correlation matrix in the self-attention neural network feature reconstruction process, and calculating the time series correlation deviation of the time series data based on the correlation matrix; wherein, based on the attention score matrix, the correlation matrix between each time step is calculated to capture the long-term and short-term dependence relationship between the time steps, i.e. the correlation matrix represents the long-term and short-term dependence relationship between the time steps, and is specifically as follows: ;​ wherein, is a correlation matrix, is an attention score matrix of an attention subnetwork in the self-attention neural network, is a number of attention subnetworks in the self-attention neural network.

[0035] Further, the Jensen-Shannon divergence (JS divergence) is used to calculate the time series correlation deviation between each sample and the normal sample, and the time series correlation deviation is specifically: ; wherein, denotes the time series correlation deviation of the sample in the time series data, denotes the JS divergence function, is an average value of the correlation matrix of all normal samples, is the correlation matrix corresponding to the sample .

[0036] Specifically, the self-attention neural network is obtained by the following way: S31, inputting the training time series sample into the neural network for processing, and calculating the time series correlation deviation loss and the reconstruction error loss of the training time series sample.

[0037] S32, based on the time series correlation deviation loss and the reconstruction error loss, calculating a training target function of the neural network, and when the training target function does not exceed the loss threshold, determining the neural network whose training target function does not exceed the loss threshold as the self-attention neural network.

[0038] wherein, during the training of the neural network, the learning rate can be set to 0.0001, the model parameters of the neural network can be updated by the Adam optimizer, the hidden layer dimension parameter of the proposed neural network is set to d=128, and the self-attention network can contain 4 layers and 8 groups of multi-head self-attention modules.

[0039] Optionally, the training target function is specifically: wherein: ; ; wherein, is a value of the training target function, denotes the reconstruction error loss, is the time series correlation deviation loss, is a weight parameter, is the time series correlation deviation of the sample in the time series data,​ For samples in time series data The reconstruction error, Represents the total number of samples in the training time series samples, Represents samples in time series data Reconstructed sequence data.

[0040] S4, based on the time series data and the corresponding reconstruction sequence data, calculate the mutual reconstruction error; wherein, The sample in the time series data The reconstruction error.

[0041] S5, using the time series correlation deviation and reconstruction error, calculates the anomaly score of the time series data and makes an anomaly judgment based on the set anomaly threshold.

[0042] The above abnormality scores are specifically as follows: ; Where, Represents samples in time series data The anomaly score of , the larger the value, the greater the possibility of anomaly in the sequence data (power load); Represents samples in time series data The reconstruction error, Represents samples in time series data The reconstructed sequence data, For samples in time series data The timing correlation deviation.

[0043] Among them, when making an abnormal judgment based on the set abnormal threshold, the abnormal threshold can be set to , if the sample Anomaly score , then the sample is judged to be abnormal.

[0044] Example 2: This embodiment of the present application provides a power load anomaly detection system based on time series correlation deviation, which is applied to a power load anomaly detection method based on time series correlation deviation in Example 1, such as Figure 4 Shown, including: A time series data acquisition module is used to collect time series data for power load anomaly detection and preprocess the time series data; The time series data reconstruction module is used to reconstruct the features of the preprocessed time series data through the self-attention neural network and obtain the corresponding reconstructed sequence data; The association deviation calculation module is configured to obtain an association matrix in a self-attention neural network feature reconstruction process, and calculate a time series association deviation of the time series data based on the association matrix. The reconstruction error calculation module is configured to calculate a reconstruction error between the time series data and the corresponding reconstructed sequence data. The anomaly score judgment module is configured to calculate an anomaly score of the time series data by using the time series association deviation and the reconstruction error, and perform anomaly judgment based on a set anomaly threshold.

[0045] In an embodiment 3, an electronic device is provided. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor implements the method of the embodiment 1 when executing the computer program.

[0046] In an embodiment 4, a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage medium stores computer instructions. The computer instructions cause a computer to execute the method of the embodiment 1.

[0047] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0048] The present application is described with reference to flowcharts and / or block diagrams according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocks Figure 1 The functions specified in a flow or multiple flows and / or blocks

[0049] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0050] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flow Figure 1 one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0051] Those skilled in the art can understand that all or part of the steps of the above-mentioned facts and methods can be completed by programs instructing the relevant hardware, and the programs involved or the programs can be stored in a computer readable storage medium. When the program is executed, the following steps are included: at this time, the corresponding method steps are derived, and the storage medium can be ROM / RAM, magnetic disc, optical disc, etc.

[0052] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting power load anomaly based on time series correlation deviation, characterized in that: The specific steps include: S1, collecting time series data for power load anomaly detection and preprocessing the time series data; S2, reconstructs the features of the preprocessed time series data through the self-attention neural network and obtains the corresponding reconstructed sequence data; S3, obtaining a correlation matrix in the self-attention neural network feature reconstruction process, and calculating the temporal correlation deviation of the time series data based on the correlation matrix; S4, calculating a mutual reconstruction error between the time series data and the corresponding reconstructed sequence data; S5, using the time series correlation deviation and the reconstruction error, calculate the anomaly score of the time series data, and perform an anomaly judgment based on a set anomaly threshold.

2. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The preprocessing includes normalization processing, specifically: ； Where, is the original time series data, ,in, is the number of samples of time series data, Indicates the number of time steps for each time series; represents the time series data after normalization. and represent the mean and standard deviation of each time series, respectively.

3. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The reconstructed sequence data is specifically: ； Where, To reconstruct sequence data, are the weight matrix and bias vector of the decoder in the self-attention neural network, Represents the output of the feedforward network in the self-attention neural network.

4. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The correlation matrix represents the long-term and short-term dependencies between each time step, specifically: ； Where, is the incidence matrix, is the attention sub-network in the self-attention neural network The attention score matrix, is the number of attention sub-networks in the self-attention neural network.

5. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The timing correlation deviation is specifically: ； Where, Represents samples in time series data The timing correlation deviation of represents the JS divergence function, is the average value of the correlation matrix of all normal samples, For samples The corresponding incidence matrix.

6. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The self-attention neural network is obtained in the following way: Input the training time series samples into the neural network for processing, and calculate the time series correlation deviation loss and reconstruction error loss of the training time series samples; Based on the temporal correlation deviation loss and the reconstruction error loss, a training objective function of the neural network is calculated. When the training objective function does not exceed a loss threshold, the neural network whose training objective function does not exceed the loss threshold is determined to be a self-attention neural network. Specifically, the training objective function is: ,in: ； ； Where, is the value of the training objective function, represents the reconstruction error loss, is the timing correlation deviation loss, is the weight parameter, For samples in time series data The timing correlation deviation of For samples in time series data The reconstruction error, Represents the total number of samples in the training time series samples, Represents samples in time series data Reconstructed sequence data.

7. The method for detecting power load anomaly based on time series correlation deviation according to claim 1, characterized in that: The abnormality score is specifically: ； Where, Represents samples in time series data The abnormality score, Represents samples in time series data The reconstruction error, Represents samples in time series data The reconstructed sequence data, For samples in time series data The timing correlation deviation.

8. A power load anomaly detection system based on time series correlation deviation, characterized in that: include: A time series data acquisition module is used to collect time series data for power load anomaly detection and preprocess the time series data; The time series data reconstruction module is used to reconstruct the features of the preprocessed time series data through the self-attention neural network and obtain the corresponding reconstructed sequence data; A correlation deviation calculation module is used to obtain a correlation matrix in the self-attention neural network feature reconstruction process, and calculate the temporal correlation deviation of the time series data based on the correlation matrix; A reconstruction error calculation module, configured to calculate a mutual reconstruction error between the time series data and the corresponding reconstruction sequence data; The anomaly score judgment module is used to calculate the anomaly score of the time series data using the time series correlation deviation and the reconstruction error, and perform anomaly judgment based on a set anomaly threshold.

9. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a method for detecting anomaly of power load based on timing correlation deviation according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions enable a computer to execute a method for detecting anomalies of power loads based on time series correlation deviation according to any one of claims 1 to 7.

Images