Power system data fusion processing method and device, storage medium and electronic equipment

Abstract

The invention discloses a power system data fusion processing method and device, a storage medium and electronic equipment. The method comprises the following steps: setting a power grid system to obtain multi-source heterogeneous data of power grid operation; calculating the multi-source heterogeneous data by setting a calculation strategy to obtain data volatility characteristics; performing anomaly detection on the multi-source heterogeneous data through a Z-score algorithm to obtain an abnormal state of the multi-source heterogeneous data; selecting a set aggregation strategy according to the data volatility characteristics and the abnormal state; performing aggregation processing on the data volatility characteristics through the set aggregation strategy to obtain aggregation characteristics; representing the aggregated features as evidences; the evidence is fused through a Dempster / Shafer evidence theory, and a merging result is obtained; and performing normalization processing on the combination result to obtain a unified quantification result. According to the method, the problem of time scale mismatching in the industrial source data use process can be effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of the use of industrial multi-source heterogeneous data in the power system, and particularly relates to a power system data fusion processing method, device, storage medium and electronic device. Background Art

[0002] With the continuous expansion of the power grid scale and the increase in operation complexity, the power grid monitoring system needs to process multi-modal data from different data sources (such as PMU, SCADA, meteorological data, etc.).

[0003] Currently, there are significant differences in the sampling frequencies and time scales of these data. For example, PMU provides high-frequency data in milliseconds, while SCADA usually samples at minute or longer intervals. This inconsistency in time scales makes it difficult to directly use the data for industrial analysis.

[0004] Therefore, it is urgent to dynamically select a suitable time scale aggregation method according to the data characteristics to uniformly process data from different sources to achieve data consistency and integrity. At the same time, to fully explore the key characteristics of the aggregated data, it is also necessary to further perform feature fusion on it to uniformly quantify and analyze the dynamic change characteristics of multi-source data. Summary of the Invention

[0005] For this reason, the present invention provides a power system data fusion processing method, device, storage medium and electronic device. By using the information contained in multi-source heterogeneous data (such as PMU, SCADA, and meteorological data, etc.), and adopting an adaptive aggregation and Dempster / Shafer evidence theory feature fusion method, it effectively solves the problem of time scale mismatch in the process of using industrial source data.

[0006] To achieve the above object, the present invention provides the following technical solutions: In the first aspect, a power system data fusion processing method is provided, including:

[0007] Obtaining multi-source heterogeneous data of power grid operation through a set power grid system;

[0008] Calculating the multi-source heterogeneous data through a set calculation strategy to obtain data volatility characteristics;

[0009] Performing anomaly detection on the multi-source heterogeneous data to obtain the anomaly state of the multi-source heterogeneous data;

[0010] Selecting a set aggregation strategy according to the data volatility characteristics and the anomaly state; aggregating the data volatility characteristics through the set aggregation strategy to obtain aggregation characteristics;

[0011] Represent the aggregation feature as evidence; fuse the evidence through Dempster / Shafer evidence theory to obtain a combined result.

[0012] As a preferred solution of the power system data fusion processing method, the data volatility feature includes volatility and change rate; the magnitude of the volatility is reflected by calculating the variance and standard deviation within a time window; the calculation formula for the variance is:

[0013]

[0014] In the formula, Var(t) is the variance; x i (t) is the i-th data point within the time window; μ(t) is the mean of the data within the window;

[0015] The calculation formula for the standard deviation is:

[0016]

[0017] In the formula, σ(t) is the standard deviation;

[0018] The calculation formula for the change rate is:

[0019]

[0020] In the formula, α is the change rate; x(t) is the data at the current time point; x(t - △t) is the data of the previous time window; △t is the time difference.

[0021] As a preferred solution of the power system data fusion processing method, in the process of performing anomaly detection on the multi-source heterogeneous data through the Z-score algorithm, the mathematical expression of the Z-score algorithm is:

[0022]

[0023] In the formula, x(t) is the data at the current time point; μ(t) is the mean of the data within this time window; σ(t) is the standard deviation.

[0024] As a preferred solution of the power system data fusion processing method, in the process of selecting the set aggregation strategy according to the data volatility feature and the abnormal state, the set aggregation strategy includes a mean aggregation strategy, a maximum aggregation strategy, a minimum aggregation strategy, and a median aggregation strategy;

[0025] The mathematical expression of the mean aggregation strategy is:

[0026]

[0027] In the formula, n is the number of data points;

[0028] The mathematical expression of the maximum value aggregation strategy is as follows:

[0029] Max(t) = max(x1(t), x2(t),..., x n (t))

[0030] The mathematical expression of the minimum value aggregation strategy is as follows:

[0031] Min(t) = min(x1(t), x2(t),..., x n (t))

[0032] The mathematical expression of the median aggregation strategy is as follows:

[0033] Median(t) = middle value(x1(t), x2(t),..., x n (t))

[0034] Where value() is the value of n data points at time t.

[0035] As a preferred solution of the power system data fusion processing method, during the process of fusing the evidence by the Dempster / Shafer evidence theory, the combination formula is:

[0036]

[0037] Where is the combined basic probability assignment, representing the confidence level of the occurrence of hypothesis A after combining evidence ε1 and ε2; is the basic probability assignment from evidence source ε1, representing the confidence level of the occurrence of hypothesis B considered by evidence source ε1; is the basic probability assignment from evidence source ε2, representing the confidence level of the occurrence of hypothesis C considered by evidence source ε2; B ∩ C = A is the intersection between hypotheses B and C from different evidence sources, that is, the part jointly supported by the two pieces of evidence; α is the adjustment factor; K is the conflict degree.

[0038] As a preferred solution of the power system data fusion processing method, after obtaining the combined result, it further includes normalizing the combined result to obtain a unified quantization result; during the process of normalizing the combined result, the normalization formula is:

[0039]

[0040] Where is the unnormalized combined result.

[0041] Second aspect, the present invention also provides a power system data fusion processing device, based on the above power system data fusion processing method, including:

[0042] A multi-source heterogeneous data acquisition module, configured to acquire multi-source heterogeneous data of power grid operation through a set power grid system;

[0043] A data volatility feature calculation module, configured to calculate the multi-source heterogeneous data through a set calculation strategy to obtain data volatility features;

[0044] An abnormal state detection module, configured to perform abnormal detection on the multi-source heterogeneous data to obtain the abnormal state of the multi-source heterogeneous data;

[0045] A feature aggregation processing module, configured to select a set aggregation strategy according to the data volatility features and the abnormal state; perform aggregation processing on the data volatility features through the set aggregation strategy to obtain aggregation features;

[0046] An evidence fusion processing module, configured to represent the aggregation features as evidence; fuse the evidence through Dempster / Shafer evidence theory to obtain a combined result.

[0047] As a preferred solution of the power system data fusion processing device, in the data volatility feature calculation module, the data volatility features include volatility and change rate; the volatility magnitude is reflected by calculating the variance and standard deviation within a time window; the calculation formula for the variance is:

[0048]

[0049] In the formula, Var(t) is the variance; x i (t) is the i-th data point within the time window; μ(t) is the mean value of the data within the window;

[0050] The calculation formula for the standard deviation is:

[0051]

[0052] In the formula, σ(t) is the standard deviation;

[0053] The calculation formula for the change rate is:

[0054]

[0055] In the formula, α is the change rate; x(t) is the data at the current time point; x(t - △t) is the data of the previous time window; △t is the time difference.

[0056] As a preferred solution of the power system data fusion processing device, in the abnormal state detection module, during the process of performing abnormal detection on the multi-source heterogeneous data through the Z-score algorithm, the mathematical expression of the Z-score algorithm is:

[0057]

[0058] In the formula, x(t) is the data at the current time point; μ(t) is the mean value of the data within this time window; σ(t) is the standard deviation.

[0059] As a preferred solution of the power system data fusion processing device, in the feature aggregation processing module, during the process of selecting the set aggregation strategy according to the data volatility characteristics and the abnormal state, the set aggregation strategy includes the mean aggregation strategy, the maximum value aggregation strategy, the minimum value aggregation strategy, and the median aggregation strategy;

[0060] The mathematical expression of the mean aggregation strategy is:

[0061]

[0062] In the formula, n is the number of data points;

[0063] The mathematical expression of the maximum value aggregation strategy is:

[0064] Max(t) = max(x1(t), x2(t),..., x n (t))

[0065] The mathematical expression of the minimum value aggregation strategy is:

[0066] Min(t) = min(x1(t), x2(t),..., x n (t))

[0067] The mathematical expression of the median aggregation strategy is:

[0068] Median(t) = middle value(x1(t), x2(t),..., x n (t))

[0069] In the formula, value() is the value of n data points at time t.

[0070] As a preferred solution of the power system data fusion processing device, in the evidence fusion processing module, during the process of fusing the evidence through the Dempster / Shafer evidence theory, the combination formula is:

[0071]

[0072] In the formula, The combined basic probability assignment represents the confidence level of the occurrence of hypothesis A after combining evidence ε1 and ε2; Is the basic probability assignment from evidence source ε1, representing the confidence level that evidence source ε1 believes hypothesis B occurs; Is the basic probability assignment from evidence source ε2, representing the confidence level that evidence source ε2 believes hypothesis C occurs; B∩C = A is the intersection between hypothesis B and C from different evidence sources, that is, the part jointly supported by the two pieces of evidence; α is the adjustment factor; K is the conflict degree.

[0073] As a preferred solution of the power system data fusion processing device, it further includes: a unified quantization result acquisition module, which is used to normalize the combined result to obtain a unified quantization result;

[0074] In the unified quantization result acquisition module, the formula for normalizing the combined result is:

[0075]

[0076] In the formula, Is the unnormalized combined result.

[0077] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, in which program codes for the power system data fusion processing method are stored, and the program codes include instructions for executing the power system data fusion processing method according to the first aspect or any possible implementation manner thereof.

[0078] In a fourth aspect, an electronic device of the present invention includes: a memory and a processor;

[0079] The processor and the memory communicate with each other through a bus; the memory stores program instructions executable by the processor, and is characterized in that the processor can execute the power system data fusion processing method according to the first aspect or any possible implementation manner thereof by calling the program instructions.

[0080] The present invention has the following advantages: The present invention obtains multi-source heterogeneous data of grid operation by setting a grid system; calculates the multi-source heterogeneous data through a set calculation strategy to obtain data volatility characteristics; performs anomaly detection on the multi-source heterogeneous data through the Z-score algorithm to obtain the anomaly state of the multi-source heterogeneous data; selects a set aggregation strategy according to the data volatility characteristics and the anomaly state; performs aggregation processing on the data volatility characteristics through the set aggregation strategy to obtain aggregation characteristics; represents the aggregation characteristics as evidence; fuses the evidence through Dempster / Shafer evidence theory to obtain a combined result; and performs normalization processing on the combined result to obtain a unified quantization result. By aiming at the volatility and change characteristics of grid data, combining with the Z-score method to detect abnormal data, and flexibly selecting the aggregation method, the present invention greatly improves the accuracy of data processing; uses Dempster / Shafer evidence theory to fuse multi-source heterogeneous data, and alleviates the influence of conflicts by introducing a regulation factor to make the state assessment more stable, and provides an intelligent fusion method for multi-source data of power systems based on soft combination Dempster / Shafer evidence theory. The present invention can first perform adaptive aggregation selection to ensure that when fusing multi-source data, the adaptive aggregation method can dynamically select an appropriate aggregation strategy (such as mean aggregation, extreme value aggregation or median aggregation) according to the change patterns of each data source (such as volatility and change rate, anomaly detection), so as to reduce unnecessary noise and data deviation, ensure the quality and consistency of the fused data, and enable grid data from different sources to obtain a more stable and meaningful integration result during fusion, laying a foundation for the next feature fusion. The soft combination technology proposed by the present invention introduces a certain degree of ambiguity or flexibility during data combination, rather than being static or fixed, and can better adapt to complex and uncertain data characteristics. Combining these two forms a new fusion framework. By introducing flexible decision-making, more efficient processing of conflicting evidence can be carried out during the combination process, reducing decision-making errors caused by conflicts, and improving the accuracy and fault tolerance of data fusion. By introducing soft combination (for example, through fuzzy logic or soft decision-making methods) to perform weighted fusion on different information sources, different evidence sources can retain their original information characteristics to a certain extent without forcing a unified decision. This synergy can improve the decision-making quality in complex and unstable environments, especially in systems such as power grids, and can better adapt to fluctuations, changes and uncertainties. The role of soft combination in resolving information conflicts and optimizing the handling of uncertainties in the application of evidence theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary. For those of ordinary skill in the art, without creative efforts, other implementation drawings can also be obtained based on the provided drawings.

[0082] The structures, ratios, sizes, etc. shown in this specification are only used to cooperate with the content disclosed in the specification for those who are familiar with this technology to understand and read, and are not used to limit the limited conditions for the implementation of the present invention. Therefore, they do not have substantial technical significance. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0083] Figure 1 It is a schematic flowchart of the power system data fusion processing method provided in Embodiment 1 of the present invention;

[0084] Figure 2 It is a schematic diagram of the architecture of the power system data fusion processing device provided in Embodiment 2 of the present invention. Specific Embodiments

[0085] The following specific embodiments illustrate the embodiments of the present invention. Those who are familiar with this technology can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0086] Embodiment 1

[0087] Refer to Figure 1 , Embodiment 1 of the present invention provides a power system data fusion processing method, including the following steps:

[0088] S1. Obtain multi-source heterogeneous data of the power grid operation through a set power grid system;

[0089] S2. Calculate the multi-source heterogeneous data through a set calculation strategy to obtain data volatility characteristics;

[0090] S3. Perform anomaly detection on the multi-source heterogeneous data through the Z-score algorithm to obtain the anomaly state of the multi-source heterogeneous data;

[0091] S4. Select a set aggregation strategy according to the data volatility characteristics and the abnormal state; perform aggregation processing on the data volatility characteristics through the set aggregation strategy to obtain aggregation characteristics;

[0092] S5. Represent the aggregation characteristics as evidence; fuse the evidence through Dempster / Shafer evidence theory to obtain a combined result;

[0093] S6. Perform normalization processing on the combined result to obtain a unified quantization result.

[0094] In this embodiment, in step S1, multi-source heterogeneous data of power grid operation is obtained through a set power grid system;

[0095] Among them, the multi-source heterogeneous data includes: node voltage, active power, reactive power, etc.

[0096] In this embodiment, in step S2, the multi-source heterogeneous data is calculated through a set calculation strategy to obtain data volatility characteristics;

[0097] Among them, the data volatility characteristics include volatility and change rate; volatility is an index to measure the degree of data change, usually represented by variance and standard deviation. The change rate represents the change rate of data over time. If the data fluctuates greatly, the change rate will be high. If the data has large volatility, we can choose to use extreme value aggregation (maximum or minimum); while if the data changes relatively smoothly, mean aggregation can be used.

[0098] The calculation formula for the variance is:

[0099]

[0100] In the formula, Var(t) is the variance; x i (t) is the i-th data point within the time window; μ(t) is the mean value of the data within the window;

[0101] The calculation formula for the standard deviation is:

[0102]

[0103] In the formula, σ(t) is the standard deviation;

[0104] The calculation formula for the change rate is:

[0105]

[0106] In the formula, α is the change rate; x(t) is the data at the current time point; x(t - △t) is the data of the previous time window; △t is the time difference.

[0107] In this embodiment, in step S3, the multi-source heterogeneous data is subjected to anomaly detection through the Z-score algorithm to obtain the anomaly status of the multi-source heterogeneous data;

[0108] Specifically, through the Z-score algorithm, it is determined whether a data point is an anomaly according to the mean and standard deviation of the data.

[0109] Among them, the mathematical expression of the Z-score algorithm is:

[0110]

[0111] In the formula, x(t) is the data at the current time point; μ(t) is the mean of the data within this time window; σ(t) is the standard deviation.

[0112] When |Z| > α (usually α = 2 or α = 3 is selected), the data can be considered an anomaly point.

[0113] In this embodiment, in step S4, according to the data volatility characteristics and the anomaly status, a set aggregation strategy is selected; the data volatility characteristics are aggregated through the set aggregation strategy to obtain aggregation characteristics;

[0114] Specifically, according to the volatility, change rate, and anomaly situation of the data, a suitable aggregation strategy is dynamically selected. Common aggregation strategies include the mean aggregation strategy, the maximum aggregation strategy, the minimum aggregation strategy, and the median aggregation strategy.

[0115] In this embodiment, if the volatility of the data is small (for example, the variance is less than a certain threshold τ var ), it indicates that the data changes smoothly, and the mean aggregation strategy is used to extract the trend within the time window. If the volatility of the data is large (for example, the variance is greater than a certain threshold τ var or the change rate is high), the maximum or minimum aggregation strategy is selected to capture the extreme changes of the data.

[0116] The mathematical expression of the mean aggregation strategy is:

[0117]

[0118] In the formula, n is the number of data points;

[0119] The maximum aggregation strategy is to select the maximum value of the data within a time window. It is applicable to capturing important events such as the maximum voltage peak and current peak in the power grid.

[0120] The mathematical expression of the maximum aggregation strategy is:

[0121] Max(t) = max(x1(t), x2(t),..., xn (t))

[0122] The minimum value aggregation strategy is to select the minimum value of the data, which is applicable to capturing situations such as the lowest voltage and current drops in the power grid.

[0123] The mathematical expression of the minimum value aggregation strategy is:

[0124] Min(t)=min(x1(t),x2(t),...,x n (t))

[0125] In some cases, extreme value aggregation may be too sensitive to outliers, resulting in distorted data processing. At this time, the median aggregation strategy can be considered. The median aggregation strategy can effectively remove the influence of abnormal data on the aggregation result. The median aggregation strategy is to calculate the median of the data within a time window, that is, the aggregation value after excluding abnormal data.

[0126] The mathematical expression of the median aggregation strategy is:

[0127] Median(t)=middle value(x1(t),x2(t),...,x n (t))

[0128] In the formula, value() is the value of n data points at time t.

[0129] In this embodiment, in step S5, the aggregated feature is represented as evidence; the evidence is fused through Dempster / Shafer evidence theory to obtain a combined result;

[0130] Specifically, in power grid monitoring, the features after adaptive aggregation processing (such as voltage volatility, active power change rate, reactive power extreme value, etc.) can be regarded as "evidence" from different data sources. These evidences reflect the dynamic changes of different monitoring indicators. Since there may be correlations and conflicts between different indicators, it is necessary to further use Dempster / Shafer evidence theory for fusion to uniformly quantify the reliability and trend of the power grid state.

[0131] The basic representation and symbol convention of evidence are as follows:

[0132] BPA (Basic Probability Assignment): m(A) is the basic probability assignment attributed to hypothesis A. Hypothesis A is a subset in the hypothesis space. Normalization condition: for all A

[0133]

[0134] Where Θ is the universal set, representing all possible hypotheses.

[0135] In this embodiment, the standard Dempster combination rule is used to combine two evidence sources m1 and m2, and the formula is as follows:

[0136]

[0137] Where K is the degree of conflict, representing the inconsistent part between evidences, and the calculation formula is:

[0138]

[0139] If K is large, it indicates that the conflict between evidence sources is relatively serious. The standard combination rule will normalize it to the part of 1 - K, excluding the conflicting part.

[0140] In this embodiment, the core idea of the soft combination Dempster / Shafer evidence theory is to control the influence of the conflict degree K on the combination result by introducing a regulation factor α. The basic formula of the soft combination rule is:

[0141]

[0142] Where is the combined basic probability assignment, representing the confidence degree that hypothesis A occurs after combining evidences ε1 and ε2; is the basic probability assignment from evidence source ε1, representing the confidence degree that evidence source ε1 believes hypothesis B occurs; is the basic probability assignment from evidence source ε2, representing the confidence degree that evidence source ε2 believes hypothesis C occurs; B ∩ C = A is the intersection between hypotheses B and C from different evidence sources, that is, the part jointly supported by the two evidences; α is the regulation factor; K is the degree of conflict.

[0143] Among them, when α is very small (close to 0), the soft combination rule is close to the standard D - S evidence combination rule, and the influence of the conflict degree K on the combination result is relatively large; when α is large, the influence of the conflict degree is weakened, and the combination result depends more on the intersection part of the evidence sources.

[0144] In this embodiment, in the soft combination, the conflict degree K reflects the conflict degree between evidence sources. Its calculation formula is:

[0145]

[0146] Where K is the situation where there is no intersection between the two evidence sources on B and C. The larger the conflict degree, the more serious the conflict between the two evidence sources.

[0147] In this embodiment, the soft combination rule can also handle uncertainty. When the conclusions given by two evidence sources are highly conflicting, the standard D-S rule will return an empty set or empty evidence, which will lead to the inability to make a reasonable decision. Soft combination "softens" this phenomenon through an adjustment factor, allowing a certain degree of uncertainty to exist. For example, if K is large in some cases, it means that the conflict is severe. Soft combination will reduce the impact of the conflict by introducing α, so that the combined result can continue to exist instead of being completely zeroed out.

[0148] In this embodiment, in the framework of soft combination, the adjustment of belief degree is also a key concept. When the conflict between evidence sources is small and reliable, the belief degree of the combined result can be increased; while when the conflict is large, the belief degree of the combined result can be decreased to reflect the uncertainty of the evidence sources. The formula for this method can be expressed as:

[0149]

[0150] This can ensure that in the case of high conflict, the result can still reasonably reflect the contribution of the evidence sources and is not affected by excessive conflicts.

[0151] In this embodiment, in step S6, the combined result is normalized to obtain a unified quantization result.

[0152] Specifically, the combined result is normalized to ensure that the sum of the probabilities after combination is 1. The normalization formula is:

[0153]

[0154] In the formula, is the unnormalized combined result. After normalization, it can ensure that the result meets the basic requirements of probability.

[0155] In a possible embodiment, the following is a specific simulation experiment example:

[0156] The feature types are divided into unaggregated features and aggregated features, the fusion methods are divided into direct Dempster / Shafer evidence theory and soft combination Dempster / Shafer evidence theory, and simulation experiments are carried out under stable and unstable states. In the following table, Dempster / Shafer is abbreviated as D-S.

[0157]

[0158] Table 1 Evaluation of the Multi-source Data Fusion Effect of Power Grid Monitoring

[0159] The experimental results are shown in Table 1. The soft combination D-S fusion method performs excellently in the stable state, with a trust degree as high as 0.91, a suspicion degree of 0.96, and extremely low uncertainty (0.05), indicating that the fusion result is highly reliable and conflicts and noises are almost eliminated. In the unstable state, the trust degree and suspicion degree drop to 0.09 and 0.14 respectively, and the uncertainty is 0.05, showing a strong conflict suppression and noise mitigation ability and significantly improving the fusion effect in the unstable situation. This method can provide higher trust degree and lower uncertainty, performs excellently in multi-source data fusion, and verifies the effectiveness of the present invention.

[0160] In summary, the present invention obtains multi-source heterogeneous data of grid operation by setting a grid system; calculates the multi-source heterogeneous data through a set calculation strategy to obtain data volatility characteristics; performs anomaly detection on the multi-source heterogeneous data through the Z-score algorithm to obtain the anomaly state of the multi-source heterogeneous data; selects a set aggregation strategy according to the data volatility characteristics and the anomaly state; performs aggregation processing on the data volatility characteristics through the set aggregation strategy to obtain aggregation characteristics; represents the aggregation characteristics as evidence; fuses the evidence through Dempster / Shafer evidence theory to obtain a combined result; and normalizes the combined result to obtain a unified quantification result. By aiming at the volatility and change characteristics of grid data, combining the Z-score method to detect abnormal data, and flexibly selecting the aggregation method, the present invention greatly improves the accuracy of data processing; uses Dempster / Shafer evidence theory to fuse multi-source heterogeneous data, and mitigates the impact of conflicts by introducing a regulation factor to make the state assessment more stable, and provides an intelligent fusion method for multi-source data of power systems based on soft combination Dempster / Shafer evidence theory. The present invention can first perform adaptive aggregation selection to ensure that when fusing multi-source data, the adaptive aggregation method can dynamically select an appropriate aggregation strategy (such as mean aggregation, extreme value aggregation, or median aggregation) according to the change patterns of each data source (such as volatility and change rate, anomaly detection), thereby reducing unnecessary noise and data deviation, ensuring the quality and consistency of the fused data, and enabling more stable and meaningful integration results for grid data from different sources, laying a foundation for the next feature fusion. The soft combination technology proposed by the present invention introduces a certain degree of ambiguity or flexibility during data combination, rather than being static or fixed, and can better adapt to complex and uncertain data characteristics. Combining these two forms a new fusion framework. By introducing flexible decision-making, more efficient processing of conflicting evidence can be achieved during the combination process, reducing decision-making errors caused by conflicts, and improving the accuracy and fault tolerance of data fusion. By introducing soft combination (for example, through fuzzy logic or soft decision-making methods) to perform weighted fusion on different information sources, different evidence sources can retain their original information characteristics to a certain extent without forcing a unified decision. This synergy can improve the decision-making quality in complex and unstable environments, especially in systems such as power grids, and can better adapt to fluctuations, changes, and uncertainties. The role of soft combination in resolving information conflicts and optimizing the handling of uncertainties in the application of evidence theory.

[0161] It should be noted that the method of the embodiments of the present disclosure can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In such a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiments of the present disclosure, and these multiple devices will interact with each other to complete the described method.

[0162] It should be noted that some embodiments of the present disclosure have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the above embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0163] Embodiment 2

[0164] See Figure 2 , Embodiment 2 of the present invention also provides a power system data fusion processing device, including:

[0165] A multi-source heterogeneous data acquisition module 001, configured to acquire multi-source heterogeneous data of power grid operation through a set power grid system;

[0166] A data volatility feature calculation module 002, configured to calculate the multi-source heterogeneous data through a set calculation strategy to obtain data volatility features;

[0167] An abnormal state detection module 003, configured to perform abnormal detection on the multi-source heterogeneous data through the Z-score algorithm to obtain the abnormal state of the multi-source heterogeneous data;

[0168] A feature aggregation processing module 004, configured to select a set aggregation strategy according to the data volatility features and the abnormal state; perform aggregation processing on the data volatility features through the set aggregation strategy to obtain aggregation features;

[0169] An evidence fusion processing module 005, configured to represent the aggregation features as evidence; fuse the evidence through Dempster / Shafer evidence theory to obtain a combined result;

[0170] A unified quantization result acquisition module 006, configured to perform normalization processing on the combined result to obtain a unified quantization result.

[0171] In this embodiment, in the data volatility feature calculation module 002, the data volatility features include volatility and change rate; the magnitude of the volatility is reflected by calculating the variance and standard deviation within a time window; the calculation formula for the variance is:

[0172]

[0173] In the formula, Var(t) is the variance; x i (t) is the i-th data point within the time window; μ(t) is the mean of the data within the window;

[0174] The calculation formula for the standard deviation is:

[0175]

[0176] In the formula, σ(t) is the standard deviation;

[0177] The calculation formula for the change rate is:

[0178]

[0179] In the formula, α is the change rate; x(t) is the data at the current time point; x(t - △t) is the data of the previous time window; △t is the time difference.

[0180] In this embodiment, in the abnormal state detection module 003, during the process of performing abnormal detection on the multi-source heterogeneous data through the Z-score algorithm, the mathematical expression of the Z-score algorithm is:

[0181]

[0182] In the formula, x(t) is the data at the current time point; μ(t) is the mean of the data within this time window; σ(t) is the standard deviation.

[0183] In this embodiment, in the feature aggregation processing module 004, during the process of selecting the set aggregation strategy according to the data volatility feature and the abnormal state, the set aggregation strategies include mean aggregation strategy, maximum aggregation strategy, minimum aggregation strategy, and median aggregation strategy;

[0184] The mathematical expression of the mean aggregation strategy is:

[0185]

[0186] In the formula, n is the number of data points;

[0187] The mathematical expression of the maximum aggregation strategy is:

[0188] Max(t) = max(x1(t), x2(t),..., x n (t))

[0189] The mathematical expression of the minimum value aggregation strategy is as follows:

[0190] Min(t) = min(x1(t), x2(t),..., x n (t))

[0191] The mathematical expression of the median value aggregation strategy is as follows:

[0192] Median(t) = middle value(x1(t), x2(t),..., x n (t))

[0193] In the formula, value() is the value of n data points at time t.

[0194] In this embodiment, in the evidence fusion processing module 005, during the process of fusing the evidence through the Dempster / Shafer evidence theory, the combination formula is as follows:

[0195]

[0196] In the formula, is the combined basic probability assignment, indicating the confidence level of the occurrence of hypothesis A after combining the evidences ε1 and ε2; is the basic probability assignment from the evidence source ε1, indicating the confidence level of the occurrence of hypothesis B considered by the evidence source ε1; is the basic probability assignment from the evidence source ε2, indicating the confidence level of the occurrence of hypothesis C considered by the evidence source ε2; B∩C = A is the intersection between the hypotheses B and C from different evidence sources, that is, the part jointly supported by the two evidences; α is the adjustment factor; K is the conflict degree.

[0197] In this embodiment, in the unified quantization result acquisition module 006, during the process of normalizing the combined result, the normalization formula is as follows:

[0198]

[0199] In the formula, is the unnormalized combined result.

[0200] It should be noted that for the information interaction, execution process, etc. between the above system modules, since they are based on the same concept as the method embodiment in Embodiment 1 of the present application, the technical effects brought by them are the same as those of the method embodiment of the present application. For the specific content, reference can be made to the description in the method embodiment shown above in the present application, and details will not be elaborated here.

[0201] Example 3

[0202] Example 3 of the present invention provides a non-transitory computer-readable storage medium, in which program codes of a power system data fusion processing method are stored, and the program codes include instructions for executing the power system data fusion processing method of Example 1 or any possible implementation manner thereof.

[0203] The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center integrating one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)), etc.

[0204] Example 4

[0205] Example 4 of the present invention provides an electronic device, including: a memory and a processor;

[0206] The processor and the memory complete communication with each other through a bus; the memory stores program instructions executable by the processor, and the processor can execute the power system data fusion processing method of Example 1 or any possible implementation manner thereof by calling the program instructions.

[0207] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor, which is implemented by reading software codes stored in the memory. The memory can be integrated in the processor or can exist independently outside the processor.

[0208] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.).

[0209] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing system. They can be concentrated on a single computing system or distributed on a network composed of multiple computing systems. Optionally, they can be implemented by program code executable by the computing system. Thus, they can be stored in the storage system and executed by the computing system. And in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules respectively, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0210] Although the present invention has been described in detail above with general descriptions and specific embodiments, on the basis of the present invention, some modifications or improvements can be made, which are obvious to those skilled in the art. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection required by the present invention.

Claims

1. A power system data fusion processing method, characterized in that: include: Acquire multi-source heterogeneous data of power grid operation by setting up the power grid system; By setting a calculation strategy to calculate the multi-source heterogeneous data, a data volatility feature is obtained; Performing anomaly detection on the multi-source heterogeneous data to obtain an abnormal state of the multi-source heterogeneous data; According to the data volatility characteristics and the abnormal state, a set aggregation strategy is selected; the data volatility characteristics are aggregated by the set aggregation strategy to obtain an aggregate feature; The aggregated features are represented as evidences; the evidences are fused through Dempster / Shafer evidence theory to obtain a merged result.

2. The power system data fusion processing method according to claim 1, characterized in that: The data volatility characteristics include volatility and rate of change; the volatility is reflected by calculating the variance and standard deviation within the time window; the calculation formula for the variance is: Where Var(t) is the variance; x i (t) is the i-th data point in the time window; μ(t) is the mean of the data in the window; The calculation formula of the standard deviation is: Where σ(t) is the standard deviation; The calculation formula of the change rate is: In the formula, α is the rate of change; x(t) is the data at the current time point; x(t-△t) is the data of the previous time window; △t is the time difference.

3. The power system data fusion processing method according to claim 2, characterized in that: In the process of performing anomaly detection on the multi-source heterogeneous data by using the Z-score algorithm, the mathematical expression of the Z-score algorithm is: Where x(t) is the data at the current time point; μ(t) is the mean of the data in the time window; σ(t) is the standard deviation.

4. The power system data fusion processing method according to claim 3 is characterized in that: In the process of selecting the set aggregation strategy according to the data volatility characteristics and the abnormal state, the set aggregation strategy includes a mean aggregation strategy, a maximum aggregation strategy, a minimum aggregation strategy and a median aggregation strategy; The mathematical expression of the mean aggregation strategy is: In the formula, n is the number of data points; The mathematical expression of the maximum aggregation strategy is: Max(t)=max(x1(t),x2(t),...,x n (t)) The mathematical expression of the minimum aggregation strategy is: Min(t)=min(x1(t),x2(t),...,x n (t)) The mathematical expression of the median aggregation strategy is: Median(t)=middle value(x1(t),x2(t),...,x n (t)) Where value() is the value of the n data points at time t.

5. The power system data fusion processing method according to claim 4, characterized in that: In the process of fusing the evidences through the Dempster / Shafer evidence theory, the merging formula is: In the formula, is the basic probability distribution after merging, which indicates the confidence level of hypothesis A after merging evidence ε1 and ε2; is the basic probability distribution from the evidence source ε1, which indicates the confidence of the evidence source ε1 that hypothesis B occurs; is the basic probability distribution from the evidence source ε2, indicating the confidence of the evidence source ε2 that hypothesis C occurs; B∩C=A is the intersection between hypotheses B and C from different evidence sources, that is, the part supported by both evidences; α is the adjustment factor; K is the degree of conflict.

6. The power system data fusion processing method according to claim 5, characterized in that: After obtaining the combined result, the method further includes normalizing the combined result to obtain a unified quantization result; in the process of normalizing the combined result, the normalization formula is: In the formula, The unnormalized merge results.

7. A power system data fusion processing device, adopting the power system data fusion processing method according to any one of claims 1 to 6, characterized in that: include: A multi-source heterogeneous data acquisition module is used to acquire multi-source heterogeneous data of power grid operation by setting a power grid system; A data volatility feature calculation module, used to calculate the multi-source heterogeneous data by setting a calculation strategy to obtain data volatility features; An abnormal state detection module, used to perform abnormality detection on the multi-source heterogeneous data to obtain the abnormal state of the multi-source heterogeneous data; A feature aggregation processing module, used for selecting and setting an aggregation strategy according to the data volatility feature and the abnormal state; performing aggregation processing on the data volatility feature by using the set aggregation strategy to obtain an aggregate feature; The evidence fusion processing module is used to represent the aggregated features as evidence; and fuse the evidence through Dempster / Shafer evidence theory to obtain a merging result.

8. The power system data fusion processing device according to claim 7, characterized in that: In the data volatility feature calculation module, the data volatility features include volatility and rate of change; the volatility is reflected by calculating the variance and standard deviation within the time window; the calculation formula of the variance is: Where Var(t) is the variance; x i (t) is the i-th data point in the time window; μ(t) is the mean of the data in the window; The calculation formula of the standard deviation is: Where σ(t) is the standard deviation; The calculation formula of the change rate is: In the formula, α is the rate of change; x(t) is the data at the current time point; x(t-△t) is the data of the previous time window; △t is the time difference.

9. The power system data fusion processing device according to claim 8, characterized in that: In the abnormal state detection module, in the process of performing abnormality detection on the multi-source heterogeneous data by using the Z-score algorithm, the mathematical expression of the Z-score algorithm is: Where x(t) is the data at the current time point; μ(t) is the mean of the data in the time window; σ(t) is the standard deviation.

10. The power system data fusion processing device according to claim 9, characterized in that: In the feature aggregation processing module, in the process of selecting the set aggregation strategy according to the data volatility characteristics and the abnormal state, the set aggregation strategy includes a mean aggregation strategy, a maximum aggregation strategy, a minimum aggregation strategy and a median aggregation strategy; The mathematical expression of the mean aggregation strategy is: In the formula, n is the number of data points; The mathematical expression of the maximum aggregation strategy is: Max(t)=max(x1(t),x2(t),...,x n (t)) The mathematical expression of the minimum aggregation strategy is: Min(t)=min(x1(t),x2(t),...,x n (t)) The mathematical expression of the median aggregation strategy is: Median(t)=middle value(x1(t),x2(t),...,x n (t)) Where value() is the value of the n data points at time t.

11. The power system data fusion processing device according to claim 10, characterized in that: In the evidence fusion processing module, the formula for fusing the evidence using the Dempster / Shafer evidence theory is: In the formula, is the basic probability distribution after merging, which indicates the confidence level of hypothesis A after merging evidence ε1 and ε2; is the basic probability distribution from the evidence source ε1, which indicates the confidence of the evidence source ε1 that hypothesis B occurs; is the basic probability distribution from the evidence source ε2, indicating the confidence of the evidence source ε2 that hypothesis C occurs; B∩C=A is the intersection between hypotheses B and C from different evidence sources, that is, the part supported by both evidences; α is the adjustment factor; K is the degree of conflict.

12. The power system data fusion processing device according to claim 11, characterized in that: Also includes: A unified quantization result acquisition module is used to normalize the combined result to obtain a unified quantization result; In the unified quantization result acquisition module, the formula for normalizing the combined result is: In the formula, The unnormalized merge results.

13. A non-transitory computer-readable storage medium, wherein the computer-readable storage medium stores a program code of a power system data fusion processing method, characterized in that: The program code includes instructions for executing the power system data fusion processing method according to any one of claims 1 to 6.

14. An electronic device, comprising: Memory and processor; The processor and the memory communicate with each other via a bus; The memory stores program instructions executable by the processor, wherein the processor calls the program instructions to execute the power system data fusion processing method according to any one of claims 1 to 6.

Images