Anomaly detection method and system for photovoltaic energy storage system

Abstract

The invention relates to the technical field of photovoltaic power generation, and provides an anomaly detection method and system for a photovoltaic energy storage system in order to solve the problem that the anomaly detection accuracy of the photovoltaic energy storage system is low due to different reasons, and the method comprises the following steps: collecting an initial information set of energy storage batteries in the photovoltaic energy storage system; preprocessing the initial information set to obtain a basic information set; performing hierarchical characteristic extraction according to the basic information set to obtain a characteristic information set; dividing the characteristic information set into condition information sets of power storage, energy storage and discharge operation; operation is executed on the coordination relation of the multiple variables, and an important characteristic set under each condition is extracted to obtain an analysis information set; establishing an operation condition model, and performing hierarchical combination on the analysis information set through the operation condition model to obtain a combination result set; performing integrated analysis on the combination result set to obtain abnormal identification information; and obtaining anomaly detection data through the anomaly identification information.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and particularly relates to an abnormal detection method for a photovoltaic energy storage system and an abnormal detection system for a photovoltaic energy storage system. Background Art

[0002] With the rapid development of renewable energy, photovoltaic energy storage systems have become an important part of achieving sustainable energy development. As the scale of photovoltaic energy storage systems continues to grow, the application scenarios are also increasing. Therefore, the abnormal detection of photovoltaic energy storage systems has become an urgent problem to be solved.

[0003] In the prior art, the abnormal detection methods for photovoltaic energy storage systems mainly include checking for high-temperature abnormal areas through an infrared thermal imager, electrical performance testing, visual inspection, cable detection, and inverter detection. However, the accuracy of these detection methods for abnormalities caused by different reasons is relatively low, and incorrect judgments may occur in complex environments, which may lead to abnormalities in the photovoltaic energy storage system and affect the entire distributed power source. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides an abnormal detection method and system for a photovoltaic energy storage system, which can improve the accuracy and reliability of abnormal detection of a photovoltaic energy storage system.

[0005] The technical solution adopted by the present invention is as follows:

[0006] An abnormal detection method for a photovoltaic energy storage system, comprising the following steps: collecting an initial information set of a storage battery in the photovoltaic energy storage system under different operating conditions, where the initial information set covers the current-voltage characteristic data and the temperature data of the storage battery; performing preprocessing on the initial information set to obtain a basic information set, wherein the preprocessing of the initial information set includes screening interference information, removing abnormal information, and coordinating multiple self-variable moments; performing hierarchical feature extraction on the basic information set to obtain a feature information set, wherein the hierarchical feature extraction includes basic feature extraction and related feature extraction of the self-variables; obtaining the operating conditions of the storage battery, and dividing the feature information set into situation information sets for charging, energy storage, and discharging operations; performing operations on the coordination relationship of multiple variables through the situation information set, and extracting an important feature set for each situation to obtain an analysis information set; building an operating condition model of the storage battery, and importing the analysis information set into the operating condition model, and performing hierarchical combination on the analysis information set to obtain a combined result set; performing integrated analysis on the combined result set to obtain abnormal recognition information, where the abnormal recognition information is used to determine the time period of the abnormality and the self-variable data associated with the abnormality based on the analysis of the feature comparison information; analyzing the detection data of the abnormal location and critical situation through the abnormal recognition information to obtain abnormal detection data.

[0007] In an embodiment of the present invention, the performing operations on the coordination relationship of multiple variables through the situation information set, and extracting an important feature set for each situation to obtain an analysis information set specifically includes: when the storage battery is in the charging situation, analyzing the variation impact generated between the voltage and the temperature of the storage battery, and proposing a coordinated accumulation amount of the voltage and the temperature of the storage battery, and selecting a charging high-temperature feature set, where the expression of the coordinated accumulation amount is:

[0008]

[0009] wherein, V(i) is the coordinated accumulation amount, U(i′) is the voltage at time i, W(i′) is the temperature at time i, is the expected value of the voltage, is the expected value of the temperature, i1 is the start time, and i n is the end time; when the storage battery is in the discharging situation, predicting the variation consistency rate of the current-voltage coordination moment of the storage battery, and proposing a consistent ratio of the current-voltage of the storage battery in the discharging situation to obtain an unconventional discharging feature set, where the expression of the consistent ratio of the current-voltage of the storage battery is:

[0010]

[0011] Wherein, Q is the current-voltage consistency ratio, ΔE is the change in current, and ΔU is the change in voltage; when the energy storage battery is in the energy storage state, the voltage of the energy storage battery and the floating degree of the hot and cold degree are collected, and the characteristics indicating self-power loss are discriminated to obtain a set of self-power loss characteristics. Among them, the expressions for the voltage of the energy storage battery and the floating degree of the hot and cold degree are:

[0012] ΔW = |W1 - W2|

[0013] ΔU = |U1 - U2|

[0014] Among them, ΔW is the floating degree of the hot and cold degree, W1 is the upper peak value of the hot and cold degree, W2 is the lower peak value of the hot and cold degree, ΔU is the floating degree of the voltage, U1 is the upper peak value of the voltage, and U2 is the lower peak value of the voltage; the set of battery high-temperature characteristics, the set of unconventional discharge characteristics, and the set of self-power loss characteristics are combined to obtain the set of parsed information.

[0015] In an embodiment of the present invention, the operation condition model of the energy storage battery is built, and the set of parsed information is imported into the operation condition model, and the set of parsed information is hierarchically combined to obtain a set of combined results, specifically including: based on the initial information set, the normal characteristics distribution boundary of all conditions is constructed through a normal function to obtain the operation condition model of the energy storage battery. Among them, the operation condition model is expressed according to the bell-shaped curve function of the normal function, and the expression of the bell-shaped curve function is:

[0016]

[0017] Among them, d(A) is the probability density of the normal function, e is the natural function, A is the characteristic vector, D is the expected vector of multi-level characteristics, Z is the covariance matrix of the characteristic vector, and T is the transpose matrix; the characteristic vectors of all conditions in the set of parsed information are transmitted to the operation condition model, and the offset and standard offset of the current characteristic value and the standard characteristic are calculated; a first peak quantity is set, and abnormal characteristics are selected and marked through the first peak quantity, offset, and standard offset; a set of combined results in multiple cases is obtained through the abnormal characteristics.

[0018] In an embodiment of the present invention, the integrated analysis of the set of combined results is performed to obtain abnormal identification information, specifically including: analyzing the set of combined results according to time series, and calculating the cumulative offset of abnormal characteristics through the change trend of characteristics. The continuation degree and importance degree of abnormal characteristics are distinguished through the cumulative offset to obtain comprehensive offset information. Among them, the expression of the comprehensive offset information is:

[0019]

[0020] where E is the accumulation offset, y j is the moment reduction ratio, A j is the j-th information point, λ A is the expectation, and k is a natural number; Set the second peak amount, collect the abnormal time period when the second peak amount is less than the comprehensive deviation information, and perform characteristic analysis at different levels on the independent variables associated with the abnormality to clarify the abnormal change form; Obtain the abnormal recognition information through the abnormal change travel and the comprehensive offset information, where the abnormal recognition information includes the time period when the abnormality occurs, the floating boundary of the abnormal characteristics, and the independent variable data related to the opposite sex characteristics.

[0021] In an embodiment of the present invention, the detection data for parsing and revealing the abnormal orientation and important situation through the abnormal recognition information to obtain abnormal detection data specifically includes: Discretize the time period of the abnormal recognition information and the offset degree of the independent variable, and stratify the offset amplitude, generation frequency, and duration of all abnormal characteristics to obtain stratified abnormal detection data, where the expression of the abnormal detection data is:

[0022]

[0023] where S is the risk coefficient, ξ1, ξ2, ξ3, and ξ4 are proportionality coefficients, H is the offset generation frequency, and G is the duration; Determine the exact position of the current abnormality in the energy storage battery through the abnormal characteristic position in the abnormal detection data to obtain position marking data; Predict the important situation in the abnormal detection data, and mark the abnormality with strong characteristic changes to obtain abnormal detection data.

[0024] In an embodiment of the present invention, the preprocessing of the initial information set to obtain the basic information set specifically includes: Traverse and screen the acquisition information of all independent variables in the initial information set, and remove the sampling values that do not conform to the change peak amount to obtain the first-order information set, where the expression of the change peak amount is:

[0025] L(i) = χ * η X + γ * λ A (i)

[0026] where L(i) is the change peak amount, η Xσ is the standard deviation at time i, and χ and γ are adjustment coefficients; sampling adjustment is performed on each group of information in the first-order information set, and time-point unification is performed on the information of multiple self-variables in the form of an average approximation during the acquisition period to obtain a collaborative information set; peak quantity identification is performed on all variables in the collaborative information set, and abnormal information far from the preset expectation is removed based on a preset change amplitude benchmark to obtain a basic information set.

[0027] In an embodiment of the present invention, extracting according to the hierarchical characteristics of the basic information set to obtain a characteristic information set specifically includes: using the basic information set to propose the expected value, variance, and maximum and minimum value changes during all sampling periods for the current-voltage and temperature data of the energy storage battery to obtain a basic characteristic information set; based on the basic characteristic information set, proposing the relevant data of all self-variables in multiple situations, where the relevant data among the current-current and temperature of the energy storage battery is expressed as:

[0028] O EUW = ρ1*U 2 + ρ2*e aW + ρ3*U*W + ρ4*E 2 + ρ5*E*U + ρ6*W*E

[0029] Among them, O EUW is the relevant data of the current-voltage and temperature of the energy storage battery, ρ1, ρ2, ρ3, ρ4, ρ5, and ρ6 are proportionality coefficients, and a is a constant; combining the basic characteristics and relevant data to obtain a characteristic information set, where the characteristic information set represents the internal change characteristics of the energy storage battery in multiple situations.

[0030] An abnormal detection system for a photovoltaic energy storage system, comprising: a collection module, which is used to collect an initial information set of an energy storage battery in the photovoltaic energy storage system under different operating conditions, and the initial information set covers the current-voltage characteristic data and the temperature data of the energy storage battery; a preprocessing module, which is used to perform preprocessing on the initial information set to obtain a basic information set, wherein the preprocessing of the initial information set includes screening interference information, removing abnormal information, and coordinating multiple self-variable moments; an extraction module, which is used to extract hierarchical characteristics according to the basic information set to obtain a characteristic information set, wherein the hierarchical characteristic extraction includes basic characteristic extraction and related characteristic extraction of the self-variables; a division module, which is used to obtain the operating conditions of the energy storage battery and divide the characteristic information set into situation information sets for power storage, energy storage, and discharge operations; an operation module, which is used to perform operations on the coordination relationship of multiple variables through the situation information set and extract an important characteristic set for each situation to obtain an analysis information set; a combination module, which is used to build an operating condition model of the energy storage battery, import the analysis information set into the operating condition model, and perform hierarchical combination on the analysis information set to obtain a combination result set; a first analysis module, which is used to perform integrated analysis on the combination result set to obtain abnormal identification information, and the abnormal identification information is used to determine the time period when the abnormality occurs and the self-variable data associated with the abnormality based on the analysis of the characteristic comparison information; a second analysis module, which is used to analyze the detection data of the abnormal location and the critical situation through the abnormal identification information to obtain abnormal detection data.

[0031] Advantages of the present invention:

[0032] By collecting the initial information set of the energy storage battery in the photovoltaic energy storage system, performing preprocessing on the initial information set to obtain a basic information set, then extracting hierarchical characteristics according to the basic information set to obtain a characteristic information set, dividing the characteristic information set into situation information sets for power storage, energy storage, and discharge operations, performing operations on the coordination relationship of multiple variables, extracting an important characteristic set for each situation to obtain an analysis information set, building an operating condition model, performing hierarchical combination on the analysis information set through the operating condition model to obtain a combination result set, performing integrated analysis on the combination result set to obtain abnormal identification information, and finally obtaining abnormal detection data through the abnormal identification information, the accuracy and reliability of abnormal detection of the photovoltaic energy storage system can be improved. Description of the drawings

[0033] Figure 1 It is a flowchart of an abnormal detection method for a photovoltaic energy storage system according to an embodiment of the present invention;

[0034] Figure 2 It is a block diagram of an anomaly detection system for a photovoltaic energy storage system according to an embodiment of the present invention. Detailed implementation manners

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0036] Figure 1 It is a flowchart of an anomaly detection method for a photovoltaic energy storage system according to an embodiment of the present invention.

[0037] As Figure 1 shown, the anomaly detection method for a photovoltaic energy storage system according to an embodiment of the present invention includes the following steps:

[0038] S1. Collect an initial information set of the energy storage battery in different operating conditions in the photovoltaic energy storage system. The initial information set includes current-voltage characteristic data and temperature data of the energy storage battery.

[0039] In an embodiment of the present invention, several sensors can be installed on the energy storage battery to view the self-variables such as the current, voltage, and temperature of the energy storage battery in real time. For example, a voltage sensor can be installed inside the energy storage battery to measure the voltage of the energy storage battery and its changes in real time, a temperature sensor can be installed on the surface of the energy storage battery to measure the temperature on the surface and inside of the energy storage battery in real time, and a current sensor can be installed to measure the magnitude and changes of the current of the energy storage battery during the charging and discharging processes in real time. Among them, these self-variables can intuitively reflect the state of the energy storage battery in various operating conditions. Therefore, the self-variables of the energy storage battery can be collected as the initial information set.

[0040] S2. Perform preprocessing on the initial information set to obtain a basic information set. The preprocessing of the initial information set includes screening interference information, removing abnormal information, and coordinating multiple self-variables at different times.

[0041] In an embodiment of the present invention, performing preprocessing on the initial information set to obtain a basic information set may specifically include: traversing and screening the acquired information of all self-variables in the initial information set, and removing the sampling values that do not conform to the variable peak amount to obtain a first-order information set. The expression of the variable peak amount is:

[0042] L(i) = χ * η X + γ * λA (i)

[0043] Among them, L(i) is the variable peak amount, η X is the standard deviation at time i, χ and γ are adjustment coefficients, λ A (i) is the expectation at time i; sampling adjustment is performed on each group of information in the first-order information set, and time-point unification is performed on the information of multiple self-variables in the form of an average approximation during the acquisition period to obtain a collaborative information set; peak amount identification is performed on all variables in the collaborative information set, and abnormal information far from the preset expectation is removed based on a preset variation amplitude benchmark to obtain a basic information set.

[0044] In an embodiment of the present invention, for the interference situation of the energy storage battery during operation, interference can be removed by varying the peak amount, and interference information greater than the upper peak value and less than the lower peak value can be removed according to the peak value of the variable peak amount, so as to more easily illustrate the current situation of the energy storage battery.

[0045] S3. Perform hierarchical feature extraction based on the basic information set to obtain a feature information set, where the hierarchical feature extraction includes basic feature extraction and related feature extraction of self-variables.

[0046] In an embodiment of the present invention, hierarchical feature extraction is performed on the basic information set to obtain a feature information set, which may specifically include: obtaining the expected value, variance, and maximum and minimum value changes during the entire sampling period for the current-voltage and temperature data of the energy storage battery through the basic information set to obtain a basic feature information set; based on the basic feature information set, obtaining the relevant data of all self-variables in various situations, where the relevant data among the current-current and temperature of the energy storage battery are expressed as:

[0047] O EUW =ρ1*U 2 +ρ2*e aW +ρ3*U*W+ρ4*E 2 +ρ5*E*U+ρ6*W*E

[0048] Among them, O EUW is the relevant data of the current-voltage and temperature of the energy storage battery, ρ1, ρ2, ρ3, ρ4, ρ5, and ρ6 are proportionality coefficients, U is the voltage of the energy storage battery, E is the current of the energy storage battery, W is the temperature of the energy storage battery, e is the natural constant, and a is a constant; combining the basic features and relevant data to obtain a feature information set, where the feature information set represents the internal variation characteristics of the energy storage battery in various situations.

[0049] S4. Obtain the operating conditions of the energy storage battery, and divide the feature information set into situation information sets for power storage, energy storage, and discharge operation.

[0050] In an embodiment of the present invention, for the abnormal characteristics of the energy storage battery under different operating conditions, a set of situation information for power storage, energy storage, and discharge operations can be established. Among them, in the power storage operation, according to the coordinated accumulation amount of the voltage and the temperature of the energy storage battery, the characteristic set with a higher battery temperature can be selected to quickly identify the abnormality in the power storage situation; in the discharge operation, the consistent change of the current and voltage of the energy storage battery can be predicted, and then the discharge characteristic set under abnormal conditions can be selected; in the energy storage operation, the self-power loss characteristic can be identified according to the temperature and the voltage fluctuation degree of the energy storage battery.

[0051] S5. Perform operations on the coordinated relationships of multiple variables through the situation information set, and extract the important characteristic sets in each situation to obtain an analysis information set.

[0052] In an embodiment of the present invention, operations are performed on the coordinated relationships of multiple variables through the situation information set, and the important characteristic sets in each situation are extracted to obtain an analysis information set. Specifically, it may include: when the energy storage battery is in the power storage situation, analyze the influence of the changes generated between the voltage and the temperature of the energy storage battery, and propose the coordinated accumulation amount of the voltage and the temperature of the energy storage battery, and select the high-temperature power storage characteristic set. Among them, the expression of the coordinated accumulation amount is:

[0053]

[0054] where V(i) is the coordinated accumulation amount, U(i′) is the voltage at time i, W(i′) is the temperature at time i, is the voltage expectation, is the temperature expectation, i1 is the start time, and i n is the end time; when the energy storage battery is in the discharge situation, predict the consistent change rate at the current-voltage coordination moment of the energy storage battery, and propose the consistent ratio of the current-voltage in the discharge situation of the energy storage battery to obtain an abnormal discharge characteristic set. Among them, the expression of the consistent ratio of the current-voltage of the energy storage battery is:

[0055]

[0056] where Q is the current-voltage consistent ratio, ΔE is the change amount of the current, and ΔU is the change amount of the voltage; when the energy storage battery is in the energy storage situation, collect the floating degrees of the voltage and the temperature of the energy storage battery, and identify the characteristics indicating self-power loss to obtain a self-power loss characteristic set. Among them, the expression of the floating degrees of the voltage and the temperature of the energy storage battery is:

[0057] ΔW = |W1 - W2|

[0058] ΔU = |U1 - U2|

[0059] Where ΔW is the floating degree of the hot and cold degree, W1 is the upper peak value of the hot and cold degree, W2 is the lower peak value of the hot and cold degree, ΔU is the floating degree of the voltage, U1 is the upper peak value of the voltage, and U2 is the lower peak value of the voltage; merge the set of energy storage high-temperature characteristics, the set of unconventional discharge characteristics, and the set of self-power-loss characteristics to obtain the set of analysis information.

[0060] S6. Build an operating condition model of the energy storage battery, import the set of analysis information into the operating condition model, and perform hierarchical combination on the set of analysis information to obtain the set of combination results.

[0061] In an embodiment of the present invention, building an operating condition model of the energy storage battery, importing the set of analysis information into the operating condition model, and performing hierarchical combination on the set of analysis information to obtain the set of combination results may specifically include: based on the initial information set, constructing the boundary of the normal characteristics distribution of all situations through the normal function to obtain the operating condition model of the energy storage battery, where the operating condition model is expressed according to the bell-shaped curve function of the normal function, and the expression of the bell-shaped curve function is:

[0062]

[0063] Where d(A) is the probability density of the normal function, e is the natural function, A is the characteristic vector, D is the expected vector of the multi-level characteristics, Z is the covariance matrix of the characteristic vector, and T is the transpose matrix; transmit the characteristic vectors of all situations in the set of analysis information to the operating condition model, and calculate the offset and standard offset of the current characteristic value and the standard characteristic; set the first peak quantity, select abnormal characteristics through the first peak quantity, offset, and standard offset and perform marking; obtain the set of combination results in multiple situations through the abnormal characteristics.

[0064] S7. Perform integrated analysis on the set of combination results to obtain abnormal identification information, and the abnormal identification information discriminates the time period of generating abnormality and the self-variable data associated with the abnormality based on the analysis of the characteristic comparison information.

[0065] In an embodiment of the present invention, performing integrated analysis on the set of combination results to obtain abnormal identification information may specifically include: performing analysis on the set of combination results according to the time sequence, and calculating the cumulative offset of the abnormal characteristics through the change trend of the characteristics, and identifying the continuation degree and importance degree of the abnormal characteristics through the cumulative offset to obtain the comprehensive offset information, where the expression of the comprehensive offset information is:

[0066]

[0067] where E is the accumulation offset, y j is the time reduction ratio, A j is the j-th information point, λ A is the expectation, k is a natural number; set the second peak quantity, collect the abnormal time period when the second peak quantity is less than the comprehensive deviation information, and perform characteristic analysis at different levels on the independent variables associated with the abnormality to clarify the abnormal change form; obtain the abnormal recognition information through the abnormal change behavior and the comprehensive offset information, where the abnormal recognition information includes the time period when the abnormality occurs, the floating boundary of the abnormal characteristics, and the independent variable data related to the abnormal characteristics.

[0068] In an embodiment of the present invention, the time reduction ratio y j can be expressed as:

[0069]

[0070] where ψ is the reduction coefficient, i k is the current time, i j is the time of the j-th point.

[0071] In an embodiment of the present invention, if the accumulation offset is greater than the set peak quantity, the abnormal time period can be collected, and the abnormal independent variables can be identified through multi-level characteristic analysis to more clearly clarify the abnormal transformation form.

[0072] Specifically, the reduction coefficient ψ can act on weighting the time when performing operations on the accumulation offset. By giving a greater ratio to the time closest to the current time, the reduction coefficient can make the system pay double attention to the abnormal characteristics generated in a shorter time period, and thus can more accurately detect the abnormality generated at this time.

[0073] S8. Analyze the abnormal orientation and the detection data of the critical situation through the abnormal recognition information to obtain the abnormal detection data.

[0074] In an embodiment of the present invention, step S8 may specifically include: discretizing the time period of the abnormal recognition information and the offset degree of the independent variable, and stratifying the offset amplitude, generation frequency, and duration of all abnormal characteristics to obtain the stratified abnormal detection data, where the expression of the abnormal detection data is:

[0075]

[0076] where S is the risk coefficient, ξ1, ξ2, ξ3, and ξ4 are proportionality coefficients, H is the generation frequency of the offset, G is the duration; determine the exact orientation of the current abnormality in the energy storage battery through the abnormal characteristic orientation in the abnormal detection data to obtain the orientation marking data; predict the critical situation in the abnormal detection data, and mark the abnormalities with strong characteristic changes to obtain the abnormal detection data.

[0077] In an embodiment of the present invention, by collecting an initial information set of a storage battery in a photovoltaic energy storage system, preprocessing the initial information set to obtain a basic information set, then performing hierarchical feature extraction according to the basic information set to obtain a feature information set, dividing the feature information set into situation information sets of power storage, energy storage, and discharge operation, performing operations on the coordination relationships of multiple variables, extracting important feature sets in each situation to obtain an analysis information set, building an operation situation model, performing hierarchical combination on the analysis information set through the operation situation model to obtain a combined result set, performing integrated analysis on the combined result set to obtain anomaly recognition information, and finally obtaining anomaly detection data through the anomaly recognition information. Thus, the accuracy and reliability of anomaly detection in the photovoltaic energy storage system can be improved.

[0078] To implement the anomaly detection method for the photovoltaic energy storage system in the above embodiment, the present invention also proposes an anomaly detection system for the photovoltaic energy storage system.

[0079] Figure 2 It is a block diagram of the anomaly detection system for the photovoltaic energy storage system in an embodiment of the present invention.

[0080] As Figure 2As shown in the figure, the abnormal detection system of the photovoltaic energy storage system according to the embodiment of the present invention includes: a collection module 100, a preprocessing module 200, an extraction module 300, a division module 400, an operation module 500, a combination module 600, a first analysis module 700, and a second analysis module 800. Among them, the collection module 100 is used to collect the initial information set of the energy storage battery in different operating conditions of the photovoltaic energy storage system. The initial information set covers the current-voltage characteristic data and the temperature data of the energy storage battery; the preprocessing module 200 is used to perform preprocessing on the initial information set to obtain the basic information set. Among them, the preprocessing of the initial information set includes screening interference information, removing abnormal information, and coordinating multiple self-variable moments; the extraction module 300 is used to extract according to the hierarchical characteristics of the basic information set to obtain the characteristic information set. Among them, the hierarchical characteristic extraction includes the extraction of the basic characteristics and the related characteristics of the self-variables; the division module 400 is used to obtain the operating conditions of the energy storage battery and divide the characteristic information set into the situation information sets of charging, energy storage, and discharging operations; the operation module 500 is used to perform operations on the coordination relationship of multiple variables through the situation information set and extract the important characteristic sets in each situation to obtain the analysis information set; the combination module 600 is used to build the operating condition model of the energy storage battery and import the analysis information set into the operating condition model to perform hierarchical combination on the analysis information set to obtain the combined result set; the first analysis module 700 is used to perform integrated analysis on the combined result set to obtain the abnormal identification information. The abnormal identification information is used to determine the time period when the abnormality occurs and the self-variable data associated with the abnormality based on the analysis of the characteristic comparison information; the second analysis module 800 is used to analyze the detection data of the abnormal position and the critical situation through the abnormal identification information to obtain the abnormal detection data.

[0081] In an embodiment of the present invention, the collection module 100 can install several sensors on the energy storage battery to view the self-variables such as the current, voltage, and temperature of the energy storage battery in real time. For example, a voltage sensor can be installed inside the energy storage battery to measure the voltage of the energy storage battery and its change in real time, a temperature sensor can be installed on the surface of the energy storage battery to measure the temperature on the surface and inside of the energy storage battery in real time, and a current sensor can be installed to measure the current magnitude and its change of the energy storage battery during the charging and discharging processes in real time. Among them, these self-variables can intuitively reflect the state of the energy storage battery in various operating conditions. Therefore, the self-variables of the energy storage battery can be collected as the initial information set.

[0082] In an embodiment of the present invention, the preprocessing module 200 is specifically used to: traverse and screen the acquisition information of all self-variables in the initial information set, and remove the sampling values that do not conform to the variable peak amount to obtain the first-order information set. The expression of the variable peak amount is:

[0083] L(i) = χ * η X + γ * λ A (i)

[0084] Wherein, L(i) is the variable peak amount, η X is the standard deviation at time i, χ and γ are adjustment coefficients, λ A (i) is the expectation at time i; Sampling adjustment is performed on each group of information in the first-order information set, and time-point unification is performed on the information of multiple self-variables in the form of an average approximation during the acquisition period to obtain a collaborative information set; Peak amount recognition is performed on all variables in the collaborative information set, and abnormal information far from the preset expectation is removed based on a preset change amplitude benchmark to obtain a basic information set.

[0085] In an embodiment of the present invention, for the interference situation of the energy storage battery during operation, interference can be removed by changing the peak amount, and interference information greater than the upper peak value and less than the lower peak value can be removed according to the peak value of the changed peak amount, so as to more easily explain the current situation of the energy storage battery.

[0086] In an embodiment of the present invention, the extraction module 300 can specifically be used to: Obtain the expected value, variance, and maximum and minimum value changes of the current-voltage and temperature data of the energy storage battery during the entire sampling period through the basic information set to obtain a basic characteristic information set; Based on the basic characteristic information set, relevant data of all self-variables in various situations are proposed, where the relevant data among the current-current, voltage, and temperature of the energy storage battery are expressed as:

[0087] O EUW = ρ1 * U 2 + ρ2 * e aW + ρ3 * U * W + ρ4 * E 2 + ρ5 * E * U + ρ6 * W * E

[0088] Wherein, O EUW is the relevant data of the current-voltage and temperature of the energy storage battery, ρ1, ρ2, ρ3, ρ4, ρ5, and ρ6 are proportionality coefficients, U is the voltage of the energy storage battery, E is the current of the energy storage battery, W is the temperature of the energy storage battery, e is the natural constant, and a is a constant; The basic characteristics and relevant data are combined to obtain a characteristic information set, where the characteristic information set represents the internal change characteristics of the energy storage battery in various situations.

[0089] In an embodiment of the present invention, for the abnormal characteristics of the energy storage battery under different operating conditions, the partitioning module 400 can construct a set of situation information for charging, energy storage, and discharging operations. Among them, in the charging operation situation, it can be analyzed according to the coordinated accumulation amount of the voltage and the degree of heat and cold of the energy storage battery, and the characteristic set with a higher battery temperature can be selected to quickly identify the abnormality in the charging situation; in the discharging operation situation, the consistent change of the current and voltage of the energy storage battery can be predicted, so as to select the discharging characteristic set under unconventional conditions; in the energy storage operation situation, the self-discharge characteristic of the electric quantity can be identified according to the degree of heat and cold and the floating degree of the voltage of the energy storage battery.

[0090] In an embodiment of the present invention, the operation module 500 is specifically configured to: when the energy storage battery is in the charging situation, analyze the change impact generated between the voltage and the degree of heat and cold of the energy storage battery, and propose the coordinated accumulation amount of the voltage and the degree of heat and cold of the energy storage battery, and select the charging high-temperature characteristic set. The expression of the coordinated accumulation amount is:

[0091]

[0092] where V(i) is the coordinated accumulation amount, U(i′) is the voltage at time i, W(i′) is the degree of heat and cold at time i, is the expected value of the voltage, is the expected value of the degree of heat and cold, i1 is the start time, and i n is the end time; when the energy storage battery is in the discharging situation, predict the consistent change rate of the current-voltage coordination moment of the energy storage battery, and propose the consistent ratio of the current-voltage of the energy storage battery in the discharging situation to obtain the unconventional discharging characteristic set. The expression of the consistent ratio of the current-voltage of the energy storage battery is:

[0093]

[0094] where Q is the current-voltage consistent ratio, ΔE is the change amount of the current, and ΔU is the change amount of the voltage; when the energy storage battery is in the energy storage situation, collect the floating degrees of the voltage and the degree of heat and cold of the energy storage battery, and identify the characteristic indicating self-discharge of the electric quantity to obtain the self-discharge characteristic set of the electric quantity. The expression of the floating degrees of the voltage and the degree of heat and cold of the energy storage battery is:

[0095] ΔW = |W1 - W2|

[0096] ΔU = |U1 - U2|

[0097] Among them, ΔW is the floating degree of the hot and cold level, W1 is the upper peak value of the hot and cold level, W2 is the lower peak value of the hot and cold level, ΔU is the floating degree of the voltage, U1 is the upper peak value of the voltage, and U2 is the lower peak value of the voltage; the set of charge storage high-temperature characteristics, the set of unconventional discharge characteristics, and the set of self-power-loss characteristics are combined to obtain the set of analysis information.

[0098] In an embodiment of the present invention, the combining module 600 can specifically be used to: based on the initial information set, construct the conventional characteristic distribution boundary of all situations through the normal function to obtain the operation situation model of the energy storage battery, where the operation situation model is expressed according to the bell-shaped curve function of the normal function, and the expression of the bell-shaped curve function is:

[0099]

[0100] Among them, d(A) is the probability density of the normal function, e is the natural function, A is the characteristic vector, D is the expected vector of the multi-level characteristics, Z is the covariance cross matrix of the characteristic vector, and T is the transpose matrix; the characteristic vectors of all situations in the analysis information set are transmitted to the operation situation model, and the offset and standard offset of the current characteristic value and the standard characteristic are calculated; the first peak quantity is set, and the abnormal characteristics are selected and marked through the first peak quantity, the offset, and the standard offset; the combination result set in multiple situations is obtained through the abnormal characteristics.

[0101] In an embodiment of the present invention, the first analysis module 700 can specifically be used to: analyze the combination result set according to the time sequence, calculate the cumulative offset of the abnormal characteristics through the change trend of the characteristics, and identify the continuation degree and importance degree of the abnormal characteristics through the cumulative offset to obtain the comprehensive offset information, where the expression of the comprehensive offset information is:

[0102]

[0103] Among them, E is the cumulative offset, y j is the time reduction ratio, A j is the jth information point, λ A is the expectation, and k is a natural number; the second peak quantity is set, and the abnormal time period when the second peak quantity is less than the comprehensive deviation information is collected, and different-level characteristic analyses are performed on the self-variables associated with the abnormality to clarify the abnormal change form; the abnormal identification information is obtained through the abnormal change trend and the comprehensive offset information, where the abnormal identification information includes the time period when the abnormality occurs, the floating boundary of the abnormal characteristics, and the self-variable data related to the opposite-sex characteristics.

[0104] In an embodiment of the present invention, the time reduction ratio y j can be expressed as:

[0105]

[0106] Among them, ψ is the reduction coefficient, and i k is the current moment, and i j is the moment of the j-th point.

[0107] In an embodiment of the present invention, if the accumulated offset is greater than the set peak amount, abnormal time periods can be collected, and abnormal self-variables can be identified through multi-level characteristic analysis to more clearly define the abnormal transformation form.

[0108] Specifically, the reduction coefficient ψ can act on weighting the moment when performing operations on the accumulated offset. By giving a larger proportion to the moment closest to the current moment, the reduction coefficient can make the system pay double attention to the abnormal characteristics generated in a shorter time period, thereby more accurately detecting the abnormalities generated at this time.

[0109] In an embodiment of the present invention, the second analysis module 800 can specifically be used to: discretize the time period of the abnormal identification information and the offset degree of the self-variable, and stratify the offset amplitude, generation frequency, and continuation duration of all abnormal characteristics to obtain stratified abnormal detection data. Among them, the expression of the abnormal detection data is:

[0110]

[0111] Among them, S is the risk coefficient, ξ1, ξ2, ξ3, and ξ4 are proportionality coefficients, H is the generation frequency of the offset, and G is the continuation moment; through the abnormal characteristic orientation in the abnormal detection data, clarify the exact orientation of the current abnormality in the energy storage battery to obtain orientation marking data; predict the critical conditions in the abnormal detection data, and mark the abnormalities with strong characteristic changes to obtain abnormal detection data.

[0112] To sum up, the present invention collects the initial information set of the energy storage battery in the photovoltaic energy storage system through the collection module, the preprocessing module preprocesses the initial information set to obtain the basic information set, then the extraction module performs hierarchical characteristic extraction according to the basic information set to obtain the characteristic information set, the division module divides the characteristic information set into the situation information sets of power storage, energy storage, and discharge operation, the operation module performs operations on the coordination relationship of various variables, the combination module extracts the important characteristic sets in each situation to obtain the analysis information set, and builds an operation situation model. The first analysis module performs hierarchical combination on the analysis information set through the operation situation model to obtain the combined result set, and performs integrated analysis on the combined result set to obtain the abnormal identification information. Finally, the second analysis module obtains the abnormal detection data through the abnormal identification information. Thus, the accuracy and reliability of abnormal detection in the photovoltaic energy storage system can be improved.

[0113] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The meaning of "a plurality" is two or more unless specifically and explicitly defined otherwise.

[0114] In the present invention, unless otherwise clearly defined and limited, terms such as "installed", "connected", "connected to", "fixed" and the like should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0115] In the present invention, unless otherwise clearly defined and limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or merely indicates that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "below" and "beneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or merely indicates that the horizontal height of the first feature is lower than that of the second feature.

[0116] In the description of this specification, the description with reference to terms such as "an embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0117] Any process or method description represented in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0118] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function, and can be embodied specifically in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of the computer-readable medium include the following: an electrical connection portion having one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then storing it in a computer memory.

[0119] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0120] Those of ordinary skill in the art can understand that all or part of the steps carried out in implementing the above-described embodiment methods can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0121] In addition, in various embodiments of the present invention, each functional unit can be integrated in a processing module, or each unit can exist physically alone, or two or more units can be integrated in one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0122] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. An abnormal detection method for a photovoltaic energy storage system, characterized in that, The steps include: Collect the initial information set of the energy storage battery in the photovoltaic energy storage system under different operating conditions, where the initial information set covers the current-voltage characteristic data and the temperature data of the energy storage battery; Perform preprocessing on the initial information set to obtain the basic information set. Among them, the preprocessing of the initial information set includes screening out interference information, removing abnormal information, and coordinating multiple self-variable moments; Perform hierarchical feature extraction according to the basic information set to obtain the feature information set. Among them, the hierarchical feature extraction includes basic feature extraction and related feature extraction of the self-variables; Obtain the operating conditions of the energy storage battery, and divide the feature information set into situation information sets for charging, energy storage, and discharging operations; Perform operations on the coordination relationship of multiple variables through the situation information set, and extract the important feature set in each situation to obtain the analysis information set; Build the operating condition model of the energy storage battery, import the analysis information set into the operating condition model, and perform hierarchical combination on the analysis information set to obtain the combined result set; Perform integrated analysis on the combined result set to obtain abnormal identification information. The abnormal identification information is used to determine the time period when the abnormality occurs and the self-variable data associated with the abnormality based on the analysis of the characteristic comparison information; Analyze the abnormal orientation and the detection data of the critical situation through the abnormal identification information to obtain the abnormal detection data.

2. The abnormal detection method of the photovoltaic energy storage system according to claim 1, wherein, The operation of performing operations on the coordination relationship of multiple variables through the situation information set and extracting the important feature set in each situation to obtain the analysis information set specifically includes: When the energy storage battery is in the charging situation, analyze the change impact generated between the voltage and the temperature of the energy storage battery, and propose the coordinated accumulation amount of the voltage and the temperature of the energy storage battery, and select the charging high-temperature feature set. Among them, the expression of the coordinated accumulation amount is: Among them, V(i) is the coordinated accumulation amount, U(i′) is the voltage at time i, and W(i′) is the degree of hot and cold at time i. is the expected value of the voltage, is the expected value of the degree of hot and cold, i1 is the start time, and i n is the end time; When the energy storage battery is in the discharging situation, predict the change coincidence rate of the current-voltage synchronization moment of the energy storage battery, and propose the coincidence ratio of the current-voltage in the discharging situation of the energy storage battery to obtain the unconventional discharging feature set. Among them, the expression of the coincidence ratio of the current-voltage of the energy storage battery is: Where Q is the current-voltage coincidence ratio, ΔE is the change amount of the current, and ΔU is the change amount of the voltage; When the energy storage battery is in the energy storage situation, collect the floating degree of the voltage and the temperature of the energy storage battery, and identify the characteristics indicating self-power loss to obtain the self-power loss feature set. Among them, the expression of the floating degree of the voltage and the temperature of the energy storage battery is: ΔW = |W1 - W2| ΔU = |U1 - U2| Where ΔW is the floating degree of the temperature, W1 is the upper peak value of the temperature, W2 is the lower peak value of the temperature, ΔU is the floating degree of the voltage, U1 is the upper peak value of the voltage, and U2 is the lower peak value of the voltage; Merge the charging high-temperature feature set, the unconventional discharging feature set, and the self-power loss feature set to obtain the analysis information set.

3. The abnormal detection method of the photovoltaic energy storage system according to claim 2, characterized in that, Construct the operation condition model of the energy storage battery, import the parsed information set into the operation condition model, and perform hierarchical combination on the parsed information set to obtain a combined result set, specifically including: Based on the initial information set, construct the conventional characteristic distribution boundary of all conditions through a normal function to obtain the operation condition model of the energy storage battery. Among them, the operation condition model is expressed according to the bell-shaped curve function of the normal function, and the expression of the bell-shaped curve function is: Where d(A) is the probability density of the normal function, e is the natural function, A is the characteristic vector, D is the expected vector of multi-level characteristics, Z is the covariance cross matrix of the characteristic vector, and T is the transpose matrix; Transfer the characteristic vectors of all conditions in the parsed information set to the operation condition model, and calculate the offset and standard offset of the current characteristic value and the standard characteristic; Set the first peak quantity, and select and label abnormal characteristics through the first peak quantity, offset and standard offset; Obtain a combined result set in multiple situations through the abnormal characteristics.

4. The abnormal detection method for a photovoltaic energy storage system according to claim 3, wherein The integrated analysis of the combined result set to obtain abnormal identification information specifically includes: Perform analysis on the combined result set according to time sequence, calculate the cumulative offset of abnormal characteristics through the change trend of characteristics, and identify the continuation degree and importance degree of abnormal characteristics through the cumulative offset to obtain comprehensive offset information. Among them, the expression of the comprehensive offset information is: Among them, E is the accumulation offset, y j is the moment reduction ratio, A j is the j-th information point, λ A is the expectation, and k is a natural number; Set the second peak quantity, collect the abnormal time periods when the second peak quantity is less than the comprehensive deviation information, and perform multi-level characteristic analysis on the independent variables associated with the abnormality to clarify the abnormal change form; Obtain abnormal identification information through the abnormal change trend and comprehensive offset information. Among them, the abnormal identification information includes the time period when the abnormality occurs, the floating boundary of abnormal characteristics, and the independent variable data related to the opposite-sex characteristics.

5. The abnormal detection method of the photovoltaic energy storage system according to claim 4, characterized in that, Analyze the detection data of the abnormal position and importance status through the abnormal identification information to obtain abnormal detection data, specifically including: Discretize the time period and the offset degree of the independent variable of the abnormal identification information, and stratify the offset amplitude, occurrence frequency and continuation duration of all abnormal characteristics to obtain stratified abnormal detection data. Among them, the expression of the abnormal detection data is: Where S is the risk coefficient, ξ1, ξ2, ξ3 and ξ4 are proportionality coefficients, H is the occurrence frequency of the offset, and G is the continuation time; Determine the exact position of the current abnormality in the energy storage battery through the abnormal characteristic position in the abnormal detection data to obtain position marking data; Predict the importance status in the abnormal detection data, and label the abnormalities with strong characteristic changes to obtain abnormal detection data.

6. The abnormal detection method of the photovoltaic energy storage system according to claim 5, characterized in that, Preprocess the initial information set to obtain a basic information set, specifically including: Traverse and screen the acquisition information of all independent variables in the initial information set, and remove the sampling values that do not meet the change peak quantity to obtain a first-order information set. Among them, the expression of the change peak quantity is: L(i) = χ * η X + γ * λ A (i) where L(i) is the variable peak volume, η X is the standard deviation at time i, and χ and γ are adjustment coefficients; Perform sampling adjustment on each group of information in the first-order information set, and perform time-point unification on the information of multiple self-variables in the form of average approximation during acquisition to obtain a collaborative information set; Perform peak volume identification on all variables in the collaborative information set, and remove abnormal information far from the preset expectation based on the preset change amplitude benchmark to obtain a basic information set.

7. The abnormal detection method of the photovoltaic energy storage system according to claim 6, wherein The extraction of the characteristic information set according to the hierarchical characteristics of the basic information set specifically includes: Propose the expected value, variance, and maximum and minimum value changes during all sampling periods for the current-voltage and temperature data of the energy storage battery through the basic information set to obtain a basic characteristic information set; Based on the basic characteristic information set, propose the relevant data of all self-variables in multiple situations, where the relevant data among the current-current and temperature of the energy storage battery are expressed as: O EUW = ρ1 * U 2 + ρ2 * e aW + ρ3 * U * W + ρ4 * E 2 + ρ5 * E * U + ρ6 * W * E Among them, O EUW is the relevant data of the current-voltage and temperature (hot and cold) of the energy storage battery, ρ1, ρ2, ρ3, ρ4, ρ5, and ρ6 are proportionality coefficients, and a is a constant; Combine the basic characteristics and relevant data to obtain a characteristic information set, where the characteristic information set represents the internal change characteristics of the energy storage battery in multiple situations.

8. An abnormal detection system for a photovoltaic energy storage system, characterized in that, Including: A collection module for collecting an initial information set of the energy storage battery in different operating conditions in the photovoltaic energy storage system, where the initial information set covers the current-voltage characteristic data and temperature data of the energy storage battery; A preprocessing module for preprocessing the initial information set to obtain a basic information set, where the preprocessing of the initial information set includes screening interference information, removing abnormal information, and coordinating the moments of multiple self-variables; An extraction module for extracting a characteristic information set according to the hierarchical characteristics of the basic information set, where the hierarchical characteristic extraction includes basic characteristic extraction and relevant characteristic extraction of the self-variables; A division module for obtaining the operating conditions of the energy storage battery and dividing the characteristic information set into situation information sets for charge storage, energy storage, and discharge operations; An operation module for performing operations on the coordination relationship of multiple variables through the situation information set and extracting important characteristic sets in each situation to obtain an analysis information set; A combination module for building an operating condition model of the energy storage battery, importing the analysis information set into the operating condition model, and performing hierarchical combination on the analysis information set to obtain a combined result set; A first analysis module for performing integrated analysis on the combined result set to obtain abnormal identification information, where the abnormal identification information is used to identify the time period of abnormality and the self-variable data associated with the abnormality based on the analysis of characteristic comparison information; A second analysis module for analyzing the detection data of the abnormal location and critical situation through the abnormal identification information to obtain abnormal detection data.

Images