Parameter detection processing method and device for primary current-limiting resonance eliminator

Abstract

A parameter detection processing method and device for a primary current-limiting resonance eliminator belong to the technical field of electrical digital data processing, and are used for analyzing humidity value change attributes of a plurality of places on the outer surface of a high-capacity constant-current element of the primary current-limiting resonance eliminator, calculating damage data factors through a plurality of dimensions, and calculating the humidity value change attributes of a plurality of places on the outer surface of the high-capacity constant-current element of the primary current-limiting resonance eliminator. Damage data can be measured through the damage data factors, the accuracy of humidity outlier measurement is efficiently improved, the error correction function of the system on disturbance of an external magnetic field is enhanced, and therefore the accuracy of humidity monitoring and subsequent management and control of a large-capacity constant-current element of the primary current-limiting resonance eliminator can be improved. The defect that in the prior art, when an LOF algorithm is used for carrying out outlier measurement on humidity data of a large-capacity constant-current element of a primary current-limiting resonance eliminator, damaged data are often mistakenly recognized as real outlier humidity data, and therefore the measurement accuracy of the humidity outlier is weakened is effectively overcome.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical digital data processing, and in particular relates to a parameter detection and processing method and device for a primary current limiting harmonic elimination device. Background Art

[0002] A primary current limiting and harmonic elimination device is a critical power system device, primarily used to address ferroresonance issues in voltage transformers in systems with ungrounded neutral points. When ferroresonance occurs in a voltage transformer, the current in its primary winding increases significantly, potentially causing high-voltage fuses to blow or even destroying the voltage transformer. To prevent these serious consequences, a primary current limiting and harmonic elimination device is installed between the neutral point and ground on the primary side of the voltage transformer.

[0003] As mentioned in the prior art solution with patent publication number "CN201118225Y", it includes a large-capacity constant current element placed in an insulating jacket, and the large-capacity constant current element is used to stabilize the current passing therethrough at a safe value. The large-capacity constant current element has certain requirements for humidity during operation, so it is very necessary to perform detection on the humidity value on the large-capacity constant current element.

[0004] When sampling and transmitting the humidity data of the large-capacity constant current element of the primary current limiting detuning device, the disturbance of the external magnetic field wave will be detrimental to the humidity data sampled and transmitted by the humidity transmitter, thereby generating damaged data.

[0005] On the other hand, the LOF algorithm can be used to detect outliers (humidity outliers) within the humidity data of the bulk constant current element of a primary current limiting and detuning device. The LOF algorithm efficiently identifies and registers outliers and is suitable for different numerical value arrangements. Because the damaged data is very anomalous, when the LOF algorithm is used to detect outliers in the humidity data of the bulk constant current element of a primary current limiting and detuning device, the damaged data is often mistaken for true outliers when performing outlier detection on the humidity data of the bulk constant current element of a primary current limiting and detuning device. This reduces the accuracy of the humidity outlier detection (outliers within the humidity data of the bulk constant current element of the primary current limiting and detuning device). Summary of the Invention

[0006] In order to solve the defects in the prior art, the present invention proposes a parameter detection and processing device and method for a primary current limiting harmonic elimination device. The present invention effectively avoids the defect in the prior art that when using the LOF algorithm to perform outlier measurement on the humidity data of the large-capacity constant current element of the primary current limiting harmonic elimination device, the damaged data is often mistaken for the real outlier humidity data, which weakens the accuracy of the measurement of humidity outliers.

[0007] The present invention utilizes the following technical solutions.

[0008] A parameter detection and processing method for a primary current limiting harmonic elimination device, comprising:

[0009] The humidity transmitter samples the humidity data of the large-capacity constant current element of the primary current limiting and detuning device and transmits it to the controller. The controller transmits the sampled humidity data of the large-capacity constant current element of the primary current limiting and detuning device to a remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal.

[0010] A method for performing an outlier determination, comprising:

[0011] Step 1: obtain the core humidity data of the large-capacity constant current component of the primary current limiting and detuning device at the same time point and the comparison humidity data of several comparison value points, construct a core humidity queue based on the core humidity data at different time points, and construct a comparison humidity queue based on the comparison humidity data at the same value point and different time points;

[0012] Step 2: Use the LOF algorithm to perform outlier detection on the core humidity queue to obtain outlier humidity data. Within the core humidity queue, use the outlier humidity data as the midpoint to divide the humidity data into several core sub-queues. Within the control humidity queue, obtain the control humidity data in the same time period as the core sub-queue to construct a control sub-queue.

[0013] Step 3, calculating the decrement cumulative value of the outlier humidity data and the numerical points in the neighborhood, and calculating the initial damaged data factor of the outlier humidity data;

[0014] Step 4: Calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period, then obtain the cumulative value of the cosine similarity and calculate the damaged data factor of the outlier humidity data;

[0015] Step 5: When the damaged data factor exceeds a predefined factor critical value, the corresponding outlier humidity data is considered damaged data.

[0016] Furthermore, in step 1, humidity data is sampled from the outer surface of the large-capacity constant current element of the primary current limiting and harmonic eliminator using a humidity transmitter, that is, five humidity transmitters are used, wherein one humidity transmitter is installed in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and harmonic eliminator, and is used to sample the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and harmonic eliminator, and the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and harmonic eliminator is used as the core humidity data; the remaining two pairs of humidity transmitters are evenly installed around the humidity transmitter in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and harmonic eliminator, each humidity transmitter in the two pairs of humidity transmitters corresponds to a reference value point, and the humidity data sampled by the two pairs of humidity transmitters are used as the reference humidity data.

[0017] Furthermore, in step 1, when the five humidity transmitters sample humidity data, all humidity transmitters sample humidity data at the same time point each time, and the core humidity data sampled at different time points are arranged in the order of their sampling time points to form a core humidity queue, and the reference humidity data sampled at different time points are arranged in the order of their sampling time points to form a reference humidity queue.

[0018] Furthermore, in step 2, after the outlier detection is performed on the core humidity queue using the LOF algorithm, the neighbors of each obtained outlier humidity data are defined. The neighbors of the outlier humidity data include a set of numerical points formed by all humidity data in the core humidity queue whose distance from the outlier humidity data is less than the set distance threshold.

[0019] Furthermore, in step 2, the method for constructing a core sub-queue by dividing a number of humidity data using the outlier humidity data as the midpoint is as follows: within the core humidity queue, obtain a number of humidity data adjacent to the outlier humidity data, obtain a number of humidity data adjacent to the outlier humidity data, and the number of humidity data adjacent to the outlier humidity data before the outlier humidity data is the same as the number of humidity data adjacent to the outlier humidity data after the outlier humidity data. The obtained number of humidity data adjacent to the outlier humidity data and the obtained number of humidity data adjacent to the outlier humidity data after the outlier humidity data are the divided humidity data, and the divided humidity data and the outlier humidity data are regarded as the core sub-queue.

[0020] Furthermore, in step 3, the equation for the initial damaged data factor is:

[0021]

[0022] In the equation, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, T p represents the number of numerical points in the neighborhood of the pth outlier humidity data when using the LOF algorithm for outlier detection, U p Represents the value of the pth outlier humidity data in the core humidity queue, U p,v Represents the value of the numerical point within the vth nearest neighbor of the pth outlier humidity data in the core humidity queue, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p *H p) represents the use of Z-score method to evaluate C p *H p Implement standardization.

[0023] Furthermore, in step 3, the equation for the sudden change can also be:

[0024] P p =BZ(C p )+BZ(H p )

[0025] In the equation, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p ) represents the use of Z-score method to evaluate C p Execution standardization, BZ(H p ) represents the use of Z-score method to evaluate H p Implement standardization.

[0026] Furthermore, in step 4, the equation for the damaged data factor is:

[0027]

[0028] In the equation: M p Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, U p Represents the pth outlier humidity data in the core humidity queue, UB p represents the mean of all control humidity data at the corresponding time point of the pth outlier humidity data in the core humidity cohort, η is a predefined non-zero positive number, I represents the number of control sub-cohorts, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

[0029] Furthermore, in step 4, the equation for the damaged data factor may also be:

[0030]

[0031] In the equation: Mp Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, I represents the number of control sub-cohorts, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

[0032] Furthermore, in step 5, if the damaged data factor of the outlier humidity data in the core humidity queue is higher than the pre-defined factor critical value, the corresponding outlier humidity data is damaged data, and all the outlier humidity data in the core humidity queue are polled and the damaged data are removed to obtain the actual outlier humidity data in the core humidity queue. The actual outlier humidity data is the detected humidity outlier value.

[0033] A parameter detection and processing device for a primary current limiting harmonic elimination device, comprising:

[0034] A humidity transmitter is provided on the outer surface of the large-capacity constant-current element of the primary current limiting and detuning device. The humidity transmitter and the 4G module are both connected to the controller. The controller is connected to a remote computer terminal in the 4G network via the 4G module. The humidity transmitter is used to sample humidity data of the large-capacity constant-current element of the primary current limiting and detuning device and transmit it to the controller. The controller is used to transmit the sampled humidity data of the large-capacity constant-current element of the primary current limiting and detuning device to the remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal.

[0035] Remote computer terminals also include:

[0036] A construction module is used to obtain core humidity data of a large-capacity constant current component of a primary current limiting and detuning device at the same time point and comparative humidity data at a plurality of comparative value points, construct a core humidity queue based on the core humidity data at different time points, and construct a comparative humidity queue based on the comparative humidity data at the same value point and at different time points;

[0037] A partitioning module is used to perform outlier detection on the core humidity queue using the LOF algorithm to obtain outlier humidity data. Within the core humidity queue, a number of humidity data are divided using the outlier humidity data as the midpoint to construct a core sub-queue. Within the control humidity queue, control humidity data in the same time period as the core sub-queue are obtained to construct a control sub-queue.

[0038] A calculation module, which is used to calculate the decrement and accumulation values of the outlier humidity data and the numerical points in the neighborhood, and calculate the initial damaged data factor of the outlier humidity data;

[0039] A damage module is used to calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period, and then obtain the cumulative value of the cosine similarity to calculate the damage data factor of the outlier humidity data;

[0040] The confirmation module is used to determine that the corresponding outlier humidity data is damaged data when the damaged data factor is higher than a pre-defined factor critical value.

[0041] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0042] By analyzing the humidity value change properties of several places on the outer surface of the large-capacity constant current element of the primary current limiting eliminator, the damage data factor is calculated in several dimensions, and the damage data can be determined through the damage data factor, which effectively improves the accuracy of humidity outlier measurement and enhances the system's error correction function for external magnetic field disturbances, thereby improving the accuracy of humidity monitoring and subsequent control of the large-capacity constant current element of the primary current limiting eliminator. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the parameter detection and processing method for the primary current limiting harmonic elimination device described in the present invention;

[0044] Figure 2 It is a partial structural diagram of the parameter detection and processing device for the primary current limiting harmonic elimination device described in the present invention. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of the present invention more clear, the following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely express the technical solutions of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0046] like Figure 1 As shown, the parameter detection and processing method for a primary current limiting harmonic elimination device according to the present invention includes:

[0047] The humidity transmitter samples the humidity data of the large-capacity constant current element of the primary current limiting and detuning device and transmits it to the controller. The controller transmits the sampled humidity data of the large-capacity constant current element of the primary current limiting and detuning device to a remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal (the humidity outliers are the outliers in the humidity data of the large-capacity constant current element of the primary current limiting and detuning device). In this way, it is possible to measure and monitor the humidity outliers in the environment of the primary current limiting and detuning device.

[0048] A method for performing outlier determination, running on a remote computer terminal, comprises:

[0049] Step 1: obtain the core humidity data of the large-capacity constant current component of the primary current limiting and detuning device at the same time point and the comparison humidity data of several comparison value points, construct a core humidity queue based on the core humidity data at different time points (the time point is the sampling time point), and construct a comparison humidity queue based on the comparison humidity data at different time points of the same value point;

[0050] In a preferred but non-limiting embodiment of the present invention, in step 1, a humidity transmitter is used to sample the humidity data transmitted from the outer surface of the large-capacity constant current element of the primary current limiting harmonic eliminator, that is, five humidity transmitters are used, where one humidity transmitter is installed in the middle of the outer surface of the large-capacity constant current element of the primary current limiting harmonic eliminator, and is used to sample the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting harmonic eliminator, and the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting harmonic eliminator is used as the core humidity data; the remaining two pairs of humidity transmitters are evenly installed around the humidity transmitter in the middle of the outer surface of the large-capacity constant current element of the primary current limiting harmonic eliminator, and the distance between the two pairs of humidity transmitters and the core humidity data transmitter can be two meters. Each humidity transmitter in the two pairs of humidity transmitters corresponds to a reference value point, and the humidity data sampled by the two pairs of humidity transmitters is used as the reference humidity data. A reference value point is reference humidity data sampled by one of the two pairs of humidity transmitters at each sampling time point. A same value point is reference humidity data sampled by the same humidity transmitter at each sampling time point.

[0051] In a preferred but non-limiting embodiment of the present invention, in step 1, when the five humidity transmitters sample humidity data, all humidity transmitters sample humidity data at the same time. The core humidity data sampled at different time points are arranged in the order of their sampling time points to form a core humidity queue. The control humidity data sampled at different time points are arranged in the order of their sampling time points to form a control humidity queue. One control value point corresponds to one control humidity queue, and two pairs of control value points correspond to two pairs of control humidity queues.

[0052] Step 2: Use the LOF algorithm to perform outlier detection on the core humidity queue to obtain outlier humidity data. Within the core humidity queue, use the outlier humidity data as the midpoint to divide the humidity data into several core sub-queues. Within the control humidity queue, obtain the control humidity data in the same time period as the core sub-queue to construct a control sub-queue.

[0053] In a preferred but non-limiting embodiment of the present invention, in step 2, after outlier detection is performed on the core humidity queue using the LOF algorithm, a neighbor is defined for each obtained outlier humidity data. The neighbor of the outlier humidity data includes a set of numerical points (these numerical points are the humidity data in the core humidity queue) formed by all humidity data in the core humidity queue whose distance from the outlier humidity data is less than a set distance threshold. The distance is the absolute value of the amount obtained by subtracting the outlier humidity from each humidity data in the core humidity queue. The set distance threshold can be determined according to specific requirements, for example, it can be 0.15 g / m 3 .

[0054] In a preferred but non-limiting embodiment of the present invention, in step 2, the method of constructing a core sub-queue by dividing a number of humidity data using the outlier humidity data as the midpoint is as follows: within the core humidity queue, obtain a number of humidity data adjacent to the outlier humidity data, obtain a number of humidity data adjacent to the outlier humidity data, and the number of humidity data adjacent to the outlier humidity data before the outlier humidity data is the same as the number of humidity data adjacent to the outlier humidity data after the outlier humidity data. The obtained number of humidity data adjacent to the outlier humidity data and the obtained number of humidity data adjacent to the outlier humidity data after the outlier humidity data are the divided humidity data, and the divided humidity data and the outlier humidity data are regarded as the core sub-queue.

[0055] Just as in the core humidity queue, the outlier humidity data is used as the midpoint, and the twenty-five humidity data adjacent to the outlier humidity data are obtained, and the twenty-five humidity data adjacent to the outlier humidity data are obtained. The twenty-five humidity data adjacent to the outlier humidity data, the twenty-five humidity data adjacent to the outlier humidity data, and the outlier humidity data are used to form a core sub-queue B1. If there are only twenty humidity data adjacent to the outlier humidity data, which is less than twenty-five, the humidity data adjacent to the outlier humidity data can be used to fill it up. That is, the number of humidity data adjacent to the outlier humidity data is selected as thirty. Similarly, if the humidity data adjacent to the outlier is not enough, the humidity data adjacent to the outlier can be used to fill it up. The corresponding time period of the core sub-queue B1 (this time period is the sampling period of the humidity data of the core sub-queue) is (u n ,u p ),u n and u pare the starting time and ending time of the period respectively, so (u n ,u p ) period (a period is a sampling period) forms a control sub-queue, and the core sub-queue B1 corresponds to two pairs of control sub-queues.

[0056] Step 3, calculating the decrement cumulative value of the outlier humidity data and the numerical points in the neighborhood, and calculating the initial damaged data factor of the outlier humidity data, wherein the initial damaged data factor is proportional to the decrement cumulative value;

[0057] In a preferred but non-limiting embodiment of the present invention, in step 3, the equation for the initial damaged data factor is:

[0058]

[0059] In the equation, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, T p represents the number of numerical points in the neighborhood of the pth outlier humidity data when using the LOF algorithm for outlier detection, U p Represents the value of the pth outlier humidity data in the core humidity queue, U p,v Represents the value of the numerical point within the vth nearest neighbor of the pth outlier humidity data in the core humidity queue, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p *H p ) represents the use of Z-score method to evaluate C p *H p Normalization is performed. The method for obtaining the numerical points within the sub-sequence where all curvatures are zero is to use the sampling time of the numerical points within the sub-sequence as the independent variable and the corresponding numerical point value as the dependent variable to apply the least squares method to obtain the fitting equation for the numerical points within the sub-sequence, and then to obtain the numerical points where the curvature of the fitting equation is zero.

[0060] In the equation Represents the cumulative value of the pth outlier humidity data in the core humidity queue and the decrement of the numerical point in the neighborhood. The higher the value of , the higher the difference between the corresponding outlier humidity data and the numerical points in its neighborhood. The more significant the corresponding outlier humidity data is, the higher the probability that it is attributed to hidden damaged data, and the higher the corresponding initial damaged data factor. p The lower it is, the lower the number of numerical points in the neighborhood of the pth outlier humidity data in the core humidity queue is, which means that the corresponding outlier humidity data is more special among the numerical points of its neighbors. Therefore, the probability that the corresponding outlier humidity data belongs to hidden damaged data is higher, and the corresponding initial damaged data factor is also higher. P in the equation p The higher the value, the more rapid the value change in the core subqueue where the corresponding outlier humidity data is located. The higher the probability that the corresponding outlier humidity data belongs to hidden damaged data, the lower the probability that it belongs to real outlier humidity data, and the higher the initial damaged data factor of the corresponding outlier humidity data.

[0061] C in the equation p The higher the value, the more rapid the change in the values of the core sub-queue where the pth outlier humidity data in the core humidity queue is located. Therefore, the sudden change of the core sub-queue where the pth outlier humidity data value point in the core humidity queue is located will be higher. p The higher it is, the faster the rhythm of value changes in the core sub-queue where the p-th outlier humidity data in the core humidity queue is located, which means that the more rapid the value changes between the value points in the core sub-queue where the p-th outlier humidity data in the core humidity queue is located, the higher the confidence level is, and the higher the sudden change of the core sub-queue where the p-th outlier humidity data in the core humidity queue is.

[0062] In a preferred but non-limiting embodiment of the present invention, in step 3, the equation for sudden change can also be:

[0063] P p =BZ(C p )+BZ(H p )

[0064] In the equation, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p ) represents the use of Z-score method to evaluate C p Execution standardization, BZ(H p ) represents the use of Z-score method to evaluate H pImplement standardization.

[0065] Step 4: Calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period. Then, obtain the cumulative value of the cosine similarity and calculate the damaged data factor of the outlier humidity data. The damaged data factor is proportional to the product of the initial damaged data factor and the inverse of the cumulative value of the cosine similarity.

[0066] The initial damage data factor of each outlier humidity data in the core humidity queue is obtained through calculation. The initial damage data factor is obtained by performing calculations based on the core humidity queue. In the specific sampled humidity queue, if a value point is damaged data, its surrounding values will also often be affected by the damaged data and form value changes, so that the surrounding value changes of the damaged value point have the same creep properties as the surrounding values of the actual outlier humidity data. And because the humidity at different locations on the large-capacity constant current element of the primary current limiting detuning device at the same time is quite close, the coherence of the humidity value changes in the core humidity queue and several control humidity queues is analyzed, and the initial damage data factor of each outlier humidity data in the core humidity queue is improved to obtain the damage data factor of each outlier humidity data.

[0067] In a preferred but non-limiting embodiment of the present invention, in step 4, the equation for the damaged data factor is:

[0068]

[0069] In the equation: M p Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, U p Represents the pth outlier humidity data in the core humidity queue, UB p represents the mean of all control humidity data at the corresponding time point of the pth outlier humidity data in the core humidity queue, and η is a predefined non-zero positive number, which is to prevent the generation of (|U p -UB p | + η) is zero, I represents the number of control sub-queues, which can be four, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

[0070] N in the equation pThe higher the value, the higher the probability that the pth outlier humidity data in the core humidity queue is damaged data, and the corresponding damaged data factor will also be higher. In the equation (|U p -UB p The higher the value of |+η), the lower the correlation between the pth outlier humidity data in the core humidity queue and the control humidity data at the same time point. This means that the reliability of the pth outlier humidity data in the core humidity queue is lower, and the corresponding outlier humidity data is more likely to be hidden damaged data. The corresponding damaged data factor will also be higher. That is, the lower the cumulative value of the cosine similarity, the lower the coherence between the p-th outlier humidity data in the core humidity queue and the control humidity data at the same time point, which means that the authenticity of the p-th outlier humidity data in the core humidity queue is lower, and the corresponding outlier humidity data is more likely to be hidden damaged data, and its corresponding damaged data factor will also be higher.

[0071] In a preferred but non-limiting embodiment of the present invention, in step 4, the equation for the damaged data factor may also be:

[0072]

[0073] In the equation: M p Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, I represents the number of control sub-queues, which can be four, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

[0074] Step 5: When the damaged data factor exceeds a predefined factor critical value, the corresponding outlier humidity data is considered damaged data.

[0075] In a preferred but non-limiting embodiment of the present invention, in step 5, if the damaged data factor of the outlier humidity data in the core humidity queue is higher than a pre-defined factor critical value, the corresponding outlier humidity data is damaged data. The factor critical value is determined according to specific requirements, such as the factor critical value can be 60%. All outlier humidity data in the core humidity queue are polled, and the damaged data are removed to obtain the actual outlier humidity data in the core humidity queue. The actual outlier humidity data is the detected humidity outlier, thereby achieving the measurement of the outlier in the humidity data of the large-capacity constant current element of the primary current limiting detuning device, thereby improving the accuracy of the measurement.

[0076] like Figure 2 As shown, the parameter detection and processing device for a primary current limiting harmonic elimination device according to the present invention includes:

[0077] A humidity transmitter is provided on the outer surface of the large-capacity constant current element of the primary current limiting and detuning device. The humidity transmitter and the 4G module are both connected to the controller. The controller is connected to the remote computer terminal in the 4G network via the 4G module. The humidity transmitter is used to sample the humidity data of the large-capacity constant current element of the primary current limiting and detuning device and transmit it to the controller. The controller is used to transmit the sampled humidity data of the large-capacity constant current element of the primary current limiting and detuning device to the remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal (the humidity outlier is the outlier in the humidity data of the large-capacity constant current element of the primary current limiting and detuning device). In this way, the outlier measurement and monitoring of the humidity in the environment of the primary current limiting and detuning device is achieved.

[0078] A construction module is used to obtain core humidity data of a large-capacity constant current component of a primary current limiting and detuning device at the same time point and comparative humidity data at a plurality of comparative value points, construct a core humidity queue based on the core humidity data at different time points, and construct a comparative humidity queue based on the comparative humidity data at the same value point and at different time points;

[0079] A partitioning module is used to perform outlier detection on the core humidity queue using the LOF algorithm to obtain outlier humidity data. Within the core humidity queue, a number of humidity data are divided using the outlier humidity data as the midpoint to construct a core sub-queue. Within the control humidity queue, control humidity data in the same time period as the core sub-queue are obtained to construct a control sub-queue.

[0080] A calculation module, which is used to calculate the decrement and accumulation values of the outlier humidity data and the numerical points in the neighborhood, and calculate the initial damaged data factor of the outlier humidity data;

[0081] A damage module is used to calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period, and then obtain the cumulative value of the cosine similarity to calculate the damage data factor of the outlier humidity data;

[0082] The confirmation module is used to determine that the corresponding outlier humidity data is damaged data when the damaged data factor is higher than a pre-defined factor critical value.

[0083] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0084] By analyzing the humidity value change properties of several places on the outer surface of the large-capacity constant current element of the primary current limiting eliminator, the damage data factor is calculated in several dimensions, and the damage data can be determined through the damage data factor, which effectively improves the accuracy of humidity outlier measurement and enhances the system's error correction function for external magnetic field disturbances, thereby improving the accuracy of humidity monitoring and subsequent control of the large-capacity constant current element of the primary current limiting eliminator.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the specific implementation methods of the present invention can still be modified or replaced with equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection space of the claims of the present invention.

Claims

1. A parameter detection and processing method for a primary current limiting harmonic elimination device, characterized in that: include: The humidity transmitter samples the humidity data of the large-capacity constant current element of the primary current limiting and detuning device and transmits it to the controller. The controller transmits the sampled humidity data of the large-capacity constant current element of the primary current limiting and detuning device to a remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal. A method for performing an outlier determination, comprising: Step 1: obtain the core humidity data of the large-capacity constant current component of the primary current limiting and detuning device at the same time point and the comparison humidity data of several comparison value points, construct a core humidity queue based on the core humidity data at different time points, and construct a comparison humidity queue based on the comparison humidity data at the same value point and different time points; Step 2: Use the LOF algorithm to perform outlier detection on the core humidity queue to obtain outlier humidity data. Within the core humidity queue, use the outlier humidity data as the midpoint to divide the humidity data into several core sub-queues. Within the control humidity queue, obtain the control humidity data in the same time period as the core sub-queue to construct a control sub-queue. Step 3, calculating the decrement cumulative value of the outlier humidity data and the numerical points in the neighborhood, and calculating the initial damaged data factor of the outlier humidity data; Step 4: Calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period, then obtain the cumulative value of the cosine similarity and calculate the damaged data factor of the outlier humidity data; Step 5: When the damaged data factor exceeds a predefined factor critical value, the corresponding outlier humidity data is considered damaged data.

2. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 1, characterized in that: In step 1, humidity data is sampled from the outer surface of the large-capacity constant current element of the primary current limiting and detuning device using a humidity transmitter, that is, five humidity transmitters are used. Here, one humidity transmitter is installed in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and detuning device, and is used to sample the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and detuning device, and the humidity data in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and detuning device is used as the core humidity data; the remaining two pairs of humidity transmitters are evenly installed around the humidity transmitter in the middle of the outer surface of the large-capacity constant current element of the primary current limiting and detuning device, and each humidity transmitter in the two pairs of humidity transmitters corresponds to a reference value point, and the humidity data sampled by the two pairs of humidity transmitters are used as the reference humidity data.

3. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 2, characterized in that: In step 1, when the five humidity transmitters sample humidity data, all humidity transmitters sample humidity data at the same time. The core humidity data sampled at different time points are arranged in the order of their sampling time points to form a core humidity queue, and the reference humidity data sampled at different time points are arranged in the order of their sampling time points to form a reference humidity queue.

4. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 3, characterized in that: In step 2, after performing outlier detection on the core humidity queue using the LOF algorithm, the nearest neighbors of each obtained outlier humidity data are defined. The nearest neighbors of the outlier humidity data include the set of numerical points formed by all humidity data in the core humidity queue whose distance from the outlier humidity data is less than the set distance threshold. In step 2, the method for constructing a core sub-queue by dividing a number of humidity data using the outlier humidity data as the midpoint is as follows: in the core humidity queue, obtain a number of humidity data adjacent to the outlier humidity data, obtain a number of humidity data adjacent to the outlier humidity data, and the number of humidity data adjacent to the outlier humidity data before the outlier humidity data is the same as the number of humidity data adjacent to the outlier humidity data after the outlier humidity data. The obtained number of humidity data adjacent to the outlier humidity data and the obtained number of humidity data adjacent to the outlier humidity data after the outlier humidity data are the divided humidity data, and the divided humidity data and the outlier humidity data are regarded as the core sub-queue.

5. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 4, characterized in that: In step 3, the equation for the starting corrupted data factor is: In the equation, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, T p represents the number of numerical points in the neighborhood of the pth outlier humidity data when using the LOF algorithm for outlier detection, U p Represents the value of the pth outlier humidity data in the core humidity queue, U p,v Represents the value of the numerical point within the vth nearest neighbor of the pth outlier humidity data in the core humidity queue, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p *H p ) represents the use of Z-score method to evaluate C p *H p Implement standardization.

6. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 5, characterized in that: In step 3, the equation for the sudden change can also be: P p =BZ(C p )+BZ(H p ) In the equation, P p represents the sudden change of the core sub-queue where the pth outlier humidity data in the core humidity queue is located, C p Represents the mean of the first-order difference values of all adjacent numerical points in the core sub-queue where the pth outlier humidity data in the core humidity queue is located, H p represents the number of numerical points with zero curvature in the subqueue where the pth outlier humidity data in the core humidity queue is located, BZ(C p ) represents the use of Z-score method to evaluate C p Execution standardization, BZ(H p ) represents the use of Z-score method to evaluate H p Implement standardization.

7. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 6, characterized in that: In step 4, the equation for the corrupted data factor is: In the equation: M p Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p Represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, U p Represents the pth outlier humidity data in the core humidity queue, UB p represents the mean of all control humidity data at the corresponding time point of the pth outlier humidity data in the core humidity cohort, η is a predefined non-zero positive number, I represents the number of control sub-cohorts, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

8. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 7, characterized in that: In step 4, the equation for the damaged data factor can also be: In the equation: M p Represents the corrupted data factor of the pth outlier humidity data in the core humidity queue, N p represents the initial corrupted data factor of the pth outlier humidity data in the core humidity queue, I represents the number of control sub-cohorts, Q p,i represents the cosine similarity between the core sub-queue where the p-th outlier humidity data is located and the i-th control sub-queue in the corresponding period, e is the Euler number, Representatives use the Z-score method to Implement standardization.

9. The parameter detection and processing method for a primary current limiting harmonic elimination device according to claim 8, characterized in that: In step 5, if the damaged data factor of the outlier humidity data in the core humidity queue is higher than the pre-defined factor critical value, the corresponding outlier humidity data is damaged data, and all the outlier humidity data in the core humidity queue are polled and the damaged data are removed to obtain the actual outlier humidity data in the core humidity queue. The actual outlier humidity data is the detected humidity outlier value.

10. A parameter detection and processing device for a primary current limiting harmonic elimination device, characterized in that: include: A humidity transmitter is provided on the outer surface of the large-capacity constant-current element of the primary current limiting and detuning device. The humidity transmitter and the 4G module are both connected to the controller. The controller is connected to a remote computer terminal in the 4G network via the 4G module. The humidity transmitter is used to sample humidity data of the large-capacity constant-current element of the primary current limiting and detuning device and transmit it to the controller. The controller is used to transmit the sampled humidity data of the large-capacity constant-current element of the primary current limiting and detuning device to the remote computer terminal to perform outlier measurement to detect humidity outliers and display the detected humidity outliers on the remote computer terminal. Remote computer terminals also include: A construction module is used to obtain core humidity data of a large-capacity constant current component of a primary current limiting and detuning device at the same time point and comparative humidity data at a plurality of comparative value points, construct a core humidity queue based on the core humidity data at different time points, and construct a comparative humidity queue based on the comparative humidity data at the same value point and at different time points; A partitioning module is used to perform outlier detection on the core humidity queue using the LOF algorithm to obtain outlier humidity data. Within the core humidity queue, a number of humidity data are divided using the outlier humidity data as the midpoint to construct a core sub-queue. Within the control humidity queue, control humidity data in the same time period as the core sub-queue are obtained to construct a control sub-queue. A calculation module, which is used to calculate the decrement and accumulation values of the outlier humidity data and the numerical points in the neighborhood, and calculate the initial damaged data factor of the outlier humidity data; A damage module is used to calculate the cosine similarity between the core sub-queue containing the outlier humidity data and each control sub-queue in the corresponding time period, and then obtain the cumulative value of the cosine similarity to calculate the damage data factor of the outlier humidity data; The confirmation module is used to determine that the corresponding outlier humidity data is damaged data when the damaged data factor is higher than a pre-defined factor critical value.

Images