Method, device and equipment for determining health degree and task reachability of broadcast transmitter

Abstract

The present application provides a method, device and equipment for determining the health degree and task reachability of a radio transmitter, and relates to the technical field of radio transmission.The present application can first mine data with higher occurrence frequency in historical monitoring data of a target transmitter, and then mine main data dimensions with higher correlation degree.Determine evaluation indexes in the mined main data dimensions, and determine the performance score of the target transmitter.Using historical data to drive index generation, quantifying correlation probability through frequent item set, and avoiding subjective differences in personnel experience.Using the correlation degree between data dimensions, mining potential impact indexes, and improving the comprehensive coverage of evaluation.The present application combines historical data mining and analytic hierarchy process to dynamically generate evaluation indexes and weights, and improves the accuracy of determining the health degree and task reachability of a radio transmitter.

Description

Technical Field

[0001] This invention relates to the field of broadcasting technology, and in particular to methods, apparatus and equipment for determining the health status and mission reachability of broadcasting transmitters. Background Technology

[0002] In the field of wireless communication, broadcast transmitters are key equipment used to convert audio or video signals (such as radio stations or television programs) into high-frequency radio waves and transmit them over wide areas via antenna systems. Broadcast transmitters are widely used in FM television backbone relay stations to achieve signal coverage and public information dissemination. The core function of a broadcast transmitter is to ensure efficient and stable signal transmission; its performance directly affects broadcast quality.

[0003] Currently, the performance evaluation of broadcast transmitters relies heavily on human experience to determine evaluation indicators. For example, the selection of evaluation indicators is often based on personnel's industry knowledge or historical case summaries, establishing a static evaluation system. Significant differences in subjective experience among personnel lead to substantial variations in the performance evaluation indicators determined by different individuals. Furthermore, different broadcast transmitters operate under different conditions; using a uniform static evaluation system lacks attention to the specific characteristics of each transmitter, ultimately resulting in poor accuracy in determining their performance. Summary of the Invention

[0004] This invention provides a method, apparatus, and equipment for determining the health status and mission reachability of broadcast transmitters, in order to solve the problem that the existing methods, which rely on human experience to determine the performance of broadcast transmitters, have poor accuracy.

[0005] In a first aspect, embodiments of the present invention provide a method for determining the health status and mission reachability of a broadcast transmitter, including:

[0006] Acquire historical monitoring data of the target transmitter; the historical monitoring data contains multiple data dimensions;

[0007] Based on the historical monitoring data, frequent itemsets are obtained according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension.

[0008] Calculate the association probability between any two data dimensions in the frequent itemset, and based on the association probability between any two data dimensions in the frequent itemset, select multiple main data dimensions.

[0009] We selected data dimensions related to transmitter performance from each master data dimension as evaluation metrics;

[0010] For each of the aforementioned evaluation indicators, the weight of each evaluation indicator is determined based on the analytic hierarchy process (AHP).

[0011] The performance of the target transmitter is determined by scoring based on each evaluation indicator and its weight; the transmitter performance includes health or mission reachability.

[0012] In one possible implementation, based on the historical monitoring data, and according to the probability of occurrence of each data dimension combination item at different times, the frequent itemset is obtained, including:

[0013] For any data dimension in historical monitoring data, the value of the data dimension is transformed by mean conversion to map it as a trend change; the trend change includes rising, falling, or remaining flat.

[0014] Treat the trend changes of various data dimensions at the same time as a single transaction and construct a transaction set;

[0015] Based on the transaction set, the probability of occurrence of the combination item of trend change in each data dimension in different transactions is obtained by dividing the number of times the combination item appears in different transactions by the total number of transactions.

[0016] Each combination of items whose probability of occurrence is greater than a preset value is considered a frequent itemset.

[0017] In one possible implementation, for any data dimension in the historical monitoring data, the values ​​of said data dimension are transformed by mean conversion to map trend changes, including:

[0018] For any data dimension X = {X1, X2, ..., X} in historical monitoring data n The mean is calculated using the following formula:

[0019]

[0020] The following formula maps data points of the same data dimension to the mean to represent trend changes:

[0021]

[0022] Where, x i Represents the i-th data point in the data dimension; μ represents the mean; This represents the tolerance threshold, indicating the data at... Fluctuations within a certain range are normal.

[0023] In one possible implementation, the association probability between any two data dimensions in the frequent itemset is calculated, and based on the association probability between any two data dimensions in the frequent itemset, multiple main data dimensions are selected, including:

[0024] For any two data dimensions in a frequent itemset, determine the number of times the trend changes of the two data dimensions occur in the same transaction;

[0025] The ratio of the number of times the trend changes of the two data dimensions occur in the same transaction to the total number of transactions is used as the correlation probability between the two data dimensions.

[0026] If the correlation probability between two data dimensions is greater than a preset value, then the two data dimensions will be used as the primary data dimensions.

[0027] In one possible implementation, the transmitter includes multiple power amplifier modules; after filtering to obtain multiple master data dimensions, it also includes:

[0028] Each main data dimension of the power amplifier module is refined to obtain the refined main data dimensions; the refinement of each main data dimension of the power amplifier module includes:

[0029] For each master data dimension of each power amplifier module, the horizontal differences between multiple modules are determined based on the differences between different power amplifier modules;

[0030] For each master data dimension of each power amplifier module, the fluctuation distance of a single module is determined based on the distance of the master data dimension from the reasonable value range;

[0031] For each main data dimension of each power amplifier module, the attenuation of a single module is determined based on the attenuation level of the main data dimension;

[0032] For each master data dimension of each power amplifier module, anomalies in single module data are determined based on whether the master data dimension exceeds a fault threshold.

[0033] In one possible implementation, the transmitter performance refers to equipment health; the step of selecting data dimensions related to transmitter performance from each master data dimension as evaluation indicators includes:

[0034] Power amplifier temperature, exciter temperature, power amplifier current, exciter current, and overall machine current are used as indicators for assessing the health of the equipment.

[0035] In one possible implementation, the transmitter performance is mission reachability; the step of selecting data dimensions related to transmitter performance from each master data dimension as evaluation metrics includes:

[0036] The power amplifier output power, exciter output power, power amplifier reflected power, and exciter reflected power are used as evaluation indicators for the achievability of equipment tasks.

[0037] Secondly, embodiments of the present invention provide a device for determining the health status and mission reachability of a broadcast transmitter, comprising:

[0038] The acquisition module is used to acquire historical monitoring data of the target transmitter; the historical monitoring data contains multiple data dimensions;

[0039] The frequent itemset acquisition module is used to obtain frequent itemsets based on the historical monitoring data and according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension.

[0040] The main dimension filtering module is used to filter multiple main data dimensions based on the calculated association probability of any two data dimensions in the frequent itemset and the association probability of any two data dimensions in the frequent itemset.

[0041] The evaluation metric filtering module is used to filter data dimensions related to transmitter performance from each master data dimension as evaluation metrics.

[0042] The weight determination module is used to determine the weight of each evaluation indicator based on the analytic hierarchy process.

[0043] The performance determination module is used to score based on various evaluation indicators and their weights to determine the performance of the target transmitter; the transmitter performance includes health or mission reachability.

[0044] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0045] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect or any possible implementation thereof.

[0046] Fifthly, embodiments of the present invention provide a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect or any possible implementation thereof.

[0047] In this embodiment of the invention, based on historical monitoring data of the target transmitter, the first step is to mine data with high frequency of occurrence in the historical data, and then to mine principal data dimensions with high correlation. From the mined principal data dimensions, evaluation indicators related to health or task reachability are determined to determine the performance score of the target transmitter. On the one hand, this replaces traditional reliance on experience, using historical data to drive indicator generation, quantifying correlation probabilities through frequent itemsets, and avoiding subjective differences due to human experience. By utilizing the correlation between data dimensions, indicators with potential impact are mined, improving the comprehensive coverage of the evaluation. On the other hand, for the actual monitoring data of the target transmitter, personalized evaluation indicators specific to the target transmitter are mined, improving the targeting of the evaluation. This invention combines historical data mining with the analytic hierarchy process (AHP) to dynamically generate evaluation indicators and weights, improving the accuracy of determining the performance of broadcast transmitters. Attached Figure Description

[0048] Figure 1 This is an application scenario diagram of the method for determining the health status and mission reachability of a broadcast transmitter provided in the embodiments of the present invention;

[0049] Figure 2 This is a flowchart illustrating the implementation of the method for determining the health status and mission reachability of a broadcast transmitter provided in this embodiment of the invention.

[0050] Figure 3 This is a schematic diagram of the device for determining the health status and mission reachability of a broadcast transmitter provided in an embodiment of the present invention;

[0051] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0052] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0053] Figure 1 This diagram illustrates an application scenario of the method for determining the health status and mission reachability of a broadcast transmitter, as provided in this embodiment of the invention. Figure 1 As shown, the structure of a broadcast transmitter is a modular system, mainly composed of a core functional section and an auxiliary monitoring section. The core functional section mainly includes an exciter module and a power amplifier module. The power amplifier module amplifies low-power signals to a high-power level to ensure broadcast coverage strength. The exciter module generates and modulates the initial carrier signal, providing the input source for the power amplifier. For example, a broadcast transmitter includes an exciter and a power amplifier.

[0054] Existing technologies suffer from low reliability in evaluation results due to strong reliance on experience, weak data support, and lack of dynamic adaptation. This invention combines historical data mining with the analytic hierarchy process (AHP) to dynamically generate evaluation indicators and weights, thereby improving the accuracy of determining the performance of broadcast transmitters.

[0055] Figure 2 This is a flowchart illustrating the implementation of the method for determining the health status and mission reachability of a broadcast transmitter, as provided in this embodiment of the invention. (Refer to...) Figure 2 This invention provides a method for determining the health status and mission reachability of a broadcast transmitter, including:

[0056] Step 201: Obtain historical monitoring data of the target transmitter; the historical monitoring data contains multiple data dimensions;

[0057] Currently, television and radio stations accumulate a large amount of monitoring data during actual operation and maintenance, especially for transmitters.

[0058] For example, transmitter monitoring data includes operational data, fault data, or maintenance data. The data volume can reach 10 million records. Data mining of the accumulated data can yield key indicators for determining the transmitter's performance. This application uses a 102.2MHz television FM backbone repeater main unit as an example to illustrate the process of determining the indicator dimensions.

[0059] For example, after acquiring historical monitoring data of the target transmitter, the process further includes: data cleaning of the historical monitoring data. For instance, data cleaning may include: removing illegal data, meaningless initial data, data from non-monitoring periods, and discrete data from discontinuous periods to obtain valid data with continuous monitoring time. The cleaned data is then formed into a new dataset and stored for use in subsequent steps. Furthermore, each piece of monitoring data may include an ID, device name, data item category, data item value, or monitoring period.

[0060] For example, historical monitoring data includes multiple data dimensions, i.e., multiple data item categories. These dimensions may include: power supply of the entire amplifier, fan status, switching method, switching mode, AB excitation, standing wave ratio (SWR), overload status, interlocking status, output power, reflected power, SWR ratio, unbalanced power, fan speed, water temperature, water flow rate, power supply status, or power supply voltage. It should be noted that historical monitoring data is typically related to the performance of the target transmitter.

[0061] Step 202: Based on the historical monitoring data, obtain frequent itemsets according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension.

[0062] It should be noted that historical monitoring data includes the values ​​of each data dimension at every moment. There are correlations between these data dimensions, specifically reflected in the related changes in their values ​​at different times. Based on the principle that higher frequency and greater probability of occurrence equate to greater data value, frequent items are extracted. For example, frequent items represent combinations of data dimension values ​​that frequently appear in historical monitoring data across different times. These combinations can contain one or more data dimensions.

[0063] In one possible implementation, based on the historical monitoring data, and according to the probability of occurrence of each data dimension combination item at different times, the frequent itemset is obtained, including:

[0064] Step 2021: For any data dimension in the historical monitoring data, the value of the data dimension is transformed by mean conversion to map it into a trend change; the trend change includes rising, falling, or remaining flat.

[0065] In some embodiments, mapping the values ​​of any data dimension in historical monitoring data to trend changes through mean transformation includes:

[0066] Step a, for any data dimension X = {X1, X2, ..., X...} in the historical monitoring data n The mean is calculated using the following formula:

[0067]

[0068] Step b involves mapping each data point on the same data dimension to the mean using the following formula to represent trend changes:

[0069]

[0070] Where, x i Represents the i-th data point in the data dimension; μ represents the mean; This represents the tolerance threshold, indicating the data at... Fluctuations within a certain range are considered normal. That is, data of the same dimension are mapped to the mean, converting the corresponding data into trend-normalized data within the current monitoring unit.

[0071] This invention performs data normalization on the cleaned data. This process eliminates interference from the magnitude of values ​​across different dimensions, allowing data from each dimension to be compared under the same standard of change. This more clearly reveals the rising and falling relationships between variables, providing a unified and comparable trend feature foundation for subsequent association rule mining. Specifically, the normalization goal is to transform the transmitter indicator data into a mean value, converting the original values ​​into trend changes, i.e., rising, falling, or flat, thereby mapping absolute values ​​to numerical change trends.

[0072] For example, after performing data normalization, the normalized data is stored in a new table and sorted in ascending order by time, serving as the basis for association rule mining.

[0073] It should be noted that association rule mining is a crucial step in data mining. After the data has been cleaned and normalized, association rule mining is performed to calculate the correlation, consistency of trend changes, and probability of occurrence among the various data collection dimensions of the transmitter. By mining the "trend direction (upward / downward / stable)" and "association patterns," the relationships between the various data collection dimensions are discovered. This is explained in detail below.

[0074] Step 2022: Treat the trend changes of each data dimension at the same time as a transaction and construct a transaction set;

[0075] For example, a set of rule-mining transactions is constructed from the normalized data. At a given time, all combinations of trends across collected dimensions are considered a single transaction. For instance, the following table shows the set of transactions from time T1 to Tn.

[0076]

[0077]

[0078] Step 2023: Based on the transaction set, divide the number of times the combination of trend changes of each data dimension appears in different transactions by the total number of transactions to obtain the probability of occurrence of the combination.

[0079] It should be noted that frequent itemsets refer to a set of conditions that frequently appear together across all transactions. For example, if "Total Machine_Output Power: Increasing; Total Machine_Reflected Power: Decreasing" appears twice in T1 and T3, then {Total Machine_Output Power: Increasing; Total Machine_Reflected Power: Decreasing} appears twice across different transactions. A minimum support (min_sup) is set, and the Apriori algorithm is used to extract high-frequency trend combinations. If an itemset is frequent, then all its subsets are also frequent. The support of candidate itemsets is calculated as follows:

[0080]

[0081] Step 2024: Select the combinations of items whose probability of occurrence is greater than the preset value as frequent itemsets.

[0082] If the support of a combination of items in the candidate itemset is greater than or equal to min_sup, then that combination is a frequent item. Frequent itemset mining identifies items that appear frequently individually or jointly. For example, the table below shows the results of frequent item mining.

[0083] Frequent itemsets [Overall Output Power: Increase; Overall Reflected Power: Decrease; Overall VSWR: Decrease; Overall Unbalanced Power P1: Decrease;] [Overall Output Power: Decreased; Overall Reflected Power: Remains Unchanged; Overall Unbalanced Power P1: Decreased;] [Actuator 2_PA voltage: decreasing; Actuator 2_PA current: decreasing] [Overall Output Power: Decreased;] [Overall System VSWR: Decreased;]

[0084] Step 203: Calculate the association probability of any two data dimensions in the frequent itemset, and based on the association probability of any two data dimensions in the frequent itemset, select multiple main data dimensions.

[0085] It should be noted that the frequent itemset contains multiple frequent items, and each frequent item contains one or more data dimensions. By exploring the correlations between data dimensions, we can further filter out data dimensions with high correlations. That is, we can eliminate data dimensions that have low correlations with other dimensions.

[0086] In one possible implementation, the association probability between any two data dimensions in the frequent itemset is calculated, and based on the association probability between any two data dimensions in the frequent itemset, multiple main data dimensions are selected, including:

[0087] Step 2031: For the trend changes of any two data dimensions in the frequent itemset, determine the number of times the trend changes of the two data dimensions occur in the same transaction.

[0088] Step 2032: The ratio of the number of times the trend changes of the two data dimensions occur in the same transaction to the total number of transactions is used as the correlation probability of the two data dimensions.

[0089] It should be noted that association rule generation is performed by generating association rules from frequent itemsets. And calculate the association probability. The specific method is as follows:

[0090]

[0091] By statistically analyzing a specific dimension within a set of frequent items, we can obtain statistics on the correlations associated with that dimension. For example, if the overall output power of the machine shows an upward trend, the data for each correlation dimension is shown in the table below:

[0092]

[0093]

[0094] Step 2033: If the correlation probability between the two data dimensions is greater than a preset value, then the two data dimensions are taken as the main data dimensions.

[0095] For example, the correlation probability between two data dimensions is the correlation probability between the changing trends of the two data dimensions.

[0096] For example, when the association probability is greater than or equal to min_conf (e.g., 0.001), the association rule is retained, that is, the two data dimensions corresponding to the association rule are retained.

[0097] In some embodiments, after analyzing the mined data and discussing with an expert group, invalid mining results are eliminated to obtain valid mining knowledge results.

[0098] The approach to data removal can be summarized as follows:

[0099] (1) Eliminate correlations, such as the correlation between the overall power and the power amplifier current when the overall power is increasing, and conversely, the correlation between the power amplifier current and the overall power. Therefore, we determine that only the overall power, which is simple and understandable in business, should be retained as the main objective.

[0100] (2) Eliminate inclusion relationships. For example, a single transmitter may contain multiple power amplifiers, which introduces the concepts of "total power" and "power of each power amplifier". In the current business phase, only the total power of the transmitter that needs to be directly concerned with the business should be retained as the main objective.

[0101] (3) Eliminate low correlation. After reviewing and analyzing the data by the expert group, it was determined that only cases with a correlation of 5% or higher will be retained in this stage. Cases with a correlation of less than 5% are low-probability events and will be eliminated in this stage.

[0102] For example, the following table shows the correlation mining results after removing data. The correlation mining results for the upward trend of the overall machine output power are used as an example.

[0103]

[0104]

[0105] Based on the data mining results, this invention comprehensively analyzes the data of various transmitters and further eliminates inclusion and correlation relationships through transmission principles and equipment principles. For example, the main dimensions are determined by power amplifier temperature, exciter temperature, power amplifier current, exciter current, and overall machine current.

[0106] In addition to the above main dimensions, the quantitative indicators can be further subdivided and expressed based on the multi-module characteristics of the transmitter.

[0107] In one possible implementation, the transmitter includes multiple power amplifier modules; after filtering to obtain multiple master data dimensions, it also includes:

[0108] Each main data dimension of the power amplifier module is refined to obtain the refined main data dimensions; the refinement of each main data dimension of the power amplifier module includes:

[0109] For each master data dimension of each power amplifier module, the horizontal differences between multiple modules are determined based on the differences between different power amplifier modules;

[0110] For each master data dimension of each power amplifier module, the fluctuation distance of a single module is determined based on the distance of the master data dimension from the reasonable value range;

[0111] For each main data dimension of each power amplifier module, the attenuation of a single module is determined based on the attenuation level of the main data dimension;

[0112] For each master data dimension of each power amplifier module, anomalies in single module data are determined based on whether the master data dimension exceeds a fault threshold.

[0113] It should be noted that the physical meaning of the detailed quantitative indicators is mainly based on conventional transmitters. One indicator dimension corresponds to multiple modules. For example, in the temperature dimension, the power amplifier temperature is a case where a transmitter contains multiple power amplifier modules, resulting in multiple power amplifier temperatures. Considering this multi-module situation (a single module can be considered a special case of multiple modules and is compatible), the following uses "power amplifier temperature" to explain these five measurement indicators.

[0114] 1. Horizontal difference between multiple modules: The horizontal difference in temperature between multiple power amplifiers. Physically, under normal circumstances, the temperature of each power amplifier module is similar. If there is a difference, it indicates that a certain power amplifier may have a problem, thus judging that there is a potential risk of module failure.

[0115] 2. Single module fluctuation distance: The distance of each power amplifier temperature from the reasonable range. Take the value of the module with the highest value: If the temperature value is far from the reasonable range but still within the limit, it also indicates that there is an abnormality. This can be used to judge whether the module has a potential failure risk.

[0116] 3. Short-term attenuation of a single module: If the temperature of any power amplifier module deteriorates linearly within a short period of time, but the absolute value does not exceed the threshold, it indicates that there is a risk of equipment failure, and the risk of failure is relatively high.

[0117] 4. Long-term temperature decay of a single power amplifier module: If the temperature of any power amplifier module deteriorates linearly over a long period of time, but the absolute value does not exceed the threshold, then there is a risk of equipment failure.

[0118] 5. Abnormal data in a single module: If the temperature value of any power amplifier module exceeds the threshold, it indicates that the device is at risk of failure, and the risk of failure is relatively high.

[0119] Regarding the above detailed results, further explanation is needed: In most cases, the "single-module fluctuation distance" can cover the significance of the "multi-module horizontal difference" indicator. That is, the "multi-module horizontal difference" indicator is somewhat redundant. However, the reason for retaining the "multi-module horizontal difference" indicator is that it is more universal and has no business attributes; it's purely from a data perspective. When we cannot accurately determine the "single-module fluctuation distance" at a certain time, there is still an indicator dimension with the same function to ensure its accuracy. In most cases, when "single-module data anomaly" occurs, the other indicators will also be covered. But why not just use the "single-module data anomaly" dimension? For the same reason above, even if it doesn't occur, the other indicators may still occur. We cannot only look at this result. At the same time, adding this indicator is also a correction item, ensuring that the final score is consistent with a naive perception; we cannot have all the alarms occurring without any difference in the algorithm results. Among them, the three indicators "multi-module horizontal differences", "single-module short-term decay", and "single-module long-term decay" have no industry attributes and are obtained purely from data analysis; "single-module fluctuation distance" and "single-module data anomaly" have industry attributes and require effective localization understanding and settings when they take effect.

[0120] Step 204: Select data dimensions related to transmitter performance from each master data dimension as evaluation indicators;

[0121] It should be noted that, due to the diverse historical monitoring data of different transmitters, not all master data dimensions obtained from data mining are suitable for evaluating transmitter performance. Therefore, it is necessary to further filter the master data dimensions obtained from data mining to logically eliminate data dimensions that are irrelevant to transmitter performance evaluation.

[0122] For example, transmitter performance includes health or mission reachability. It should be noted that health and mission reachability are two different dimensions for evaluating transmitter performance. Health primarily refers to the transmitter's equipment status. Mission reachability primarily refers to the transmitter's ability to complete its mission.

[0123] For example, data dimensions related to transmitter health are selected from each master data dimension as health assessment indicators; correspondingly, in steps 205 and 206, the weights of each health assessment indicator are determined based on the analytic hierarchy process; and scores are calculated based on each health assessment indicator and its weights to determine the health of the target transmitter.

[0124] For example, data dimensions related to the transmitter's mission reachability are selected from each master data dimension as evaluation indicators for mission reachability. Accordingly, in steps 205 and 206, the weights of each evaluation indicator for mission reachability are determined based on the analytic hierarchy process (AHP); scores are then calculated based on each evaluation indicator and its weights to determine the mission reachability of the target transmitter.

[0125] In one possible implementation, the transmitter performance is the device health status; the step of filtering data dimensions related to transmitter performance from each master data dimension as evaluation indicators includes using power amplifier temperature, exciter temperature, power amplifier current, exciter current, and overall current as evaluation indicators for device health status.

[0126] In one possible implementation, the transmitter performance is the mission reachability; the step of filtering data dimensions related to transmitter performance from each master data dimension as evaluation indicators includes using power amplifier output power, exciter output power, power amplifier reflection power, and exciter reflection power as evaluation indicators of the device's mission reachability.

[0127] Determining the dimensions of the evaluation indicators is one of the most important tasks in the evaluation methodology and represents a core innovative achievement. The main body of the evaluation indicators is obtained through data mining of transmitter operational data; additionally, other indicator dimensions can be added from the perspectives of domain-specific professional knowledge and business experience. The indicator dimensions derived through data mining belong to the actual equipment perception data collected during transmitter operation. In addition to "knowledge mining," the perception data also includes "time-sensitive data," which has a significant impact on actual operations. Time-sensitive data is a straightforward indicator dimension. Within the transmitter's existing information infrastructure, data is collected in real-time through protocol communication with the transmitter and used for monitoring. This data includes the output power and reflection power data of two key components: the power amplifier and the exciter. For example, time-sensitive data includes output power and reflection power; output power includes the power amplifier output power and the exciter output power; reflection power includes the power amplifier reflection power and the exciter reflection power.

[0128] Regarding the indicator system obtained from the above data mining, the following further explanation is provided. The indicators include "causal relationships." If the output power does not meet expectations, it may be due to amplifier aging. In this case, the amplifier typically exhibits symptoms such as high module temperature and high module current. However, a hot amplifier module and high current do not necessarily lead to a decrease in output power. Here, amplifier aging is the cause (direct cause), the hot amplifier module is also a cause (conductive cause), and the high module current is also a cause (conductive cause), while the substandard output power is the effect. This cause-and-effect relationship can be applied to both overall output power and the output power of a single amplifier.

[0129] Currently, the "effects" of key dimensions (such as output power and transmission power) can be collected; while the "causes" of these dimensions—the "direct causes"—are certainly unavailable, the "transferred causes" can usually be collected. This is a benefit of the overall digitalization improvement in the industry. The "cause" part is suitable for guiding "health assessment," while the "effect" part is suitable for guiding "accessibility assessment." The "effect" part should also be considered for participation in health assessment, mainly to add a strong decision-making dimension, aiming to maintain consistency with basic perception. The "cause" part can be excluded from accessibility assessment, also to maintain consistency with basic perception and avoid calculating low scores while the machine is still outputting reasonably. Through data mining, the most prominent "transferred causes" are "temperature" and "current." "Timely data" only considers two indicators because this is the direction of the "effect," focusing not only on direct results but also on long-term changes to identify potential problems.

[0130] The following explanation adds evaluation indicators from other aspects such as domain-specific professional knowledge theory and industry experience summary.

[0131] In terms of domain expertise theory, the perceived data is directly collected from the transmitter. This data is acquired from a business operations perspective as the industry evolves, aiming to assist equipment maintenance and operation personnel. However, this data dimension itself suffers from survivorship bias. Adding a new indicator dimension based on the fundamental principles of domain knowledge is an effective supplement to the evaluation. Through summarizing the principles of wireless transmission and discussions with domain experts, it was ultimately decided to add "energy conversion efficiency" as a new evaluation indicator dimension. This indicator is novel and requires a calculation definition. Wireless transmission essentially involves converting electrical energy into electromagnetic wave energy. A healthy transmitter possesses energy conversion capabilities consistent with theoretical calculations. As the machine operates for extended periods, the conversion efficiency decreases, resulting in increased power consumption for the same transmission power. Therefore, "energy conversion efficiency" is defined in the algorithm as "transmission power / power consumption," that is, the ratio of the transmitter's transmission power to its power consumption.

[0132] In terms of industry experience, the wireless transmission industry has accumulated a wealth of operational experience in actual operation, including operation and maintenance, repair and maintenance, equipment management, and problem handling. Taking transmitter health as the analytical target, this study summarizes daily operation and maintenance experience, resulting in relevant indicator dimensions, primarily including: management and support experience, maintenance completion rate (completion of regular maintenance and repair work), emergency preparedness (good spare parts inventory for rapid response and resumption of broadcasting), maintenance and repair experience, manufacturer brand differentiation (considering product quality, after-sales service, technical support, and response time), impact of component damage (damage may not necessarily affect the current transmitter results, but it can cause continuous damage to other components and lead to overall capability failure), operational failure rate (statistical results of actual equipment operation data), and key criteria (feeder connector temperature: when a key criterion occurs, an anomaly is highly probable).

[0133] Other external influencing factors can also be considered. For example, ambient temperature and humidity during operation; key safeguards include power supply voltage stability (unstable voltage will lead to unstable output power), signal source system reliability, antenna and feeder system reliability, and grounding resistance effectiveness. Algorithm correction indicators should include a reserved interface for manual correction. A veto condition is defined for abnormal output power and abnormal reflected power.

[0134] Step 205: For each of the evaluation indicators, determine the indicator weights of each evaluation indicator based on the analytic hierarchy process.

[0135] This invention employs the Analytic Hierarchy Process (AHP) to determine the weights of each indicator dimension. AHP is a quantitative analysis method for multi-criteria decision-making, aiming to help people make reasonable and accurate choices in complex decision-making environments. The core idea of ​​AHP is to hierarchically structure a complex decision problem and establish a complete structural system of criteria and sub-criteria by comparing the importance of pairwise factors. This system includes a criterion layer, a sub-criteria layer, and an alternative layer, helping decision-makers analyze and compare the weight relationships between different factors. In practical applications, it is necessary to accurately construct the judgment matrix and perform consistency checks to improve the reliability and feasibility of the decision results.

[0136] For example, the factors in the first-layer AHP matrix include: mined data, timely data, operational experience, industry principles, or algorithm corrections. Taking the first-layer mined data as an example, the factors in the second-layer AHP matrix include: power amplifier temperature, exciter temperature, power amplifier current, exciter current, or overall system current. Taking the second-layer temperature or current factors as an example, the factors in the third-layer AHP matrix include: horizontal differences between multiple modules, fluctuation distance of a single module, short-term attenuation of a single module, long-term attenuation of a single module, or abnormal data in a single module.

[0137] For example, after constructing the AHP judgment matrix, the process includes: organizing a group of business experts and domain experts to perform weighted scoring according to the AHP implementation method. The scoring uses the 9-scale method commonly used in AHP to represent the comparison scale of the importance between two factors. The specific scales are 1 (equally important), 3 (slightly important), 5 (significantly important), 7 (strongly important), and 9 (extremely important). Even-numbered scales are the median values ​​of adjacent judgments, and reciprocals of each other indicate opposite importance.

[0138] For example, after weight scoring, the process includes: performing a consistency check and obtaining the weight system. According to the AHP implementation method, if the consistency check passes, the matrix and weight calculation are deemed valid, and the corresponding weight results of each element in each level are obtained, thus obtaining the index weight value.

[0139] The consistency check first calculates the consistency ratio (CR). If CR < 0.1, the matrix consistency check passes; otherwise, the judgment matrix and the consistency ratio CR need to be readjusted until it passes. The algorithm for calculating CR is as follows: Where RI is the consistency index, λmax is the largest eigenvalue of the relative importance judgment matrix, and n is the order of the relative importance judgment matrix.

[0140] Step 206: Scoring is performed based on each evaluation index and its weight to determine the performance of the target transmitter.

[0141] For example, the performance of the target transmitter is determined by a weighted scoring method based on each evaluation index and its weight.

[0142] In the weighted scoring method, each evaluation indicator corresponds to a weight, and the sum of all weights is 100%. Each evaluation indicator has its own scoring model, with a maximum score of 100 points, constructing a scoring system that includes both qualitative and quantitative (positive processing and standardized processing) dimensions. Different target transmitters are scored out of 100 points; a higher score indicates better health and a better ability to complete the mission. The score calculation formula is...

[0143]

[0144] Where: Wi is the weight of the i-th evaluation index. Si is the score of the i-th evaluation indicator, with a maximum score of 100. n is the number of evaluation indicators.

[0145] The weighted scoring model chosen aligns with the current state of the broadcast transmitter industry. This invention presents a novel method for assessing transmitter health and mission achievability, employing an intuitive and concise model as its defining principle. This method allows for simultaneous review of the validity of indicator dimensions and targeted accumulation of operational data after implementation, laying the foundation for adjusting indicator dimensions and optimizing the algorithm model.

[0146] It should be noted that when using a weighted scoring model, in addition to the indicator weights, it is also necessary to determine the scoring criteria and assign actual values ​​to the evaluation indicators. In the numerical type category, different indicator items correspond to different numerical types, including numeric and enumerated types. Each indicator item needs to have a corresponding scoring system constructed, with a minimum score of 0 and a maximum score of 100. In specific operations, the following processing methods can be used for different data types:

[0147] 1. Numerical type:

[0148] (1) Continuous values: such as temperature and voltage values. For this type, based on a reasonable value range, normalization is performed using the following formula to obtain a score. In actual calculations, if the collected values ​​are negative, they need to be "positiveed" first before calculation.

[0149]

[0150] (2) Discrete numerical values: The same processing method is used as for the following "enumeration type".

[0151] 2. Enumeration type: such as transmitter manufacturer brand differentiation, influence of human evaluation, etc. This type of scoring uses qualitative methods to obtain quantitative scores; then, through AHP, the weights of each enumerated element are obtained, and these weights are used as the quantitative scores. According to the AHP principle, the sum of these weights is 100.

[0152] 3. Boolean type: such as maintenance completion rate, whether there is any component damage, etc. These indicators are assigned values ​​by switching signals. If the maintenance is completed, a value of 100 is assigned; otherwise, a value of 0 is assigned.

[0153] In one possible implementation, after determining the performance of the target transmitter by scoring it based on various evaluation indicators and their weights, the method further includes: determining the degree of failure risk of the target transmitter based on its performance score. It should be noted that the performance score of the target transmitter indicates the probability that the target transmitter itself may fail; therefore, the degree of failure risk of the target transmitter can be monitored in real time based on its performance score.

[0154] In this embodiment of the invention, based on historical monitoring data of the target transmitter, the process first mines data with high frequency of occurrence, and then mines principal data dimensions with high correlation. Evaluation indicators are determined from the mined principal data dimensions to determine the performance score of the target transmitter. On the one hand, this replaces traditional reliance on experience, using historical data to drive indicator generation and quantifying correlation probabilities through frequent itemsets, avoiding subjective differences based on human experience. By utilizing the correlation between data dimensions, indicators with potential impact are mined, improving the comprehensiveness of the evaluation. On the other hand, for the actual monitoring data of the target transmitter, personalized evaluation indicators specific to the target transmitter are mined, improving the targeting of the evaluation. This invention combines historical data mining with the analytic hierarchy process (AHP) to dynamically generate evaluation indicators and weights, improving the accuracy of determining the performance of broadcast transmitters.

[0155] Existing methods for target assessment using the analytic hierarchy process (AHP) often rely on static data and expert experience to construct their judgment matrices, lacking historical data support and leading to issues with the validity of the results. The key difference in this application is that its judgment matrix is ​​constructed entirely based on historical data. Through data mining of historical data, it scientifically constructs the indicator system required for the judgment matrix.

[0156] Due to differences in operational phases, station conditions, and transmitter characteristics, the scoring benchmarks calculated for different transmitters may vary. Furthermore, the dimensions of the calculated indicators may also change as operational effectiveness analysis progresses. For example, the exciter temperature indicator is currently only collected on a few transmitters, and its impact on data mining is significant; this indicator may not be suitable for other transmitters. The solution implements dynamic configuration functionality, supporting dynamic configuration of indicator dimensions and indicator score (numerical type), allowing for real-time optimization of method execution performance.

[0157] This invention combines historical data mining (frequent itemsets) with the Analytic Hierarchy Process (AHP) to dynamically generate evaluation indicators and weights, thereby achieving precise quantification of the performance of broadcast transmitters (e.g., health and mission reachability).

[0158] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0159] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0160] Figure 3 A schematic diagram of the device for determining the health status and mission reachability of a broadcast transmitter according to an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below:

[0161] like Figure 3 As shown, a device 3 for determining the health status and mission reachability of a broadcast transmitter includes:

[0162] The acquisition module 31 is used to acquire historical monitoring data of the target transmitter; the historical monitoring data contains multiple data dimensions;

[0163] The frequent itemset acquisition module 32 is used to obtain frequent itemsets based on the historical monitoring data and according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension.

[0164] The main dimension filtering module 33 is used to calculate the association probability of any two data dimensions in the frequent itemset, and to filter out multiple main data dimensions based on the association probability of any two data dimensions in the frequent itemset.

[0165] The evaluation index screening module 34 is used to screen data dimensions related to transmitter performance from each master data dimension as evaluation indicators.

[0166] The weight determination module 35 is used to determine the weight of each evaluation indicator based on the analytic hierarchy process.

[0167] The performance determination module 36 is used to score based on each evaluation index and its weight to determine the performance of the target transmitter.

[0168] In this embodiment of the invention, based on historical monitoring data of the target transmitter, the process first mines data with high frequency of occurrence, and then mines principal data dimensions with high correlation. Evaluation indicators are determined from the mined principal data dimensions to determine the performance score of the target transmitter. On the one hand, this replaces traditional reliance on experience, using historical data to drive indicator generation and quantifying correlation probabilities through frequent itemsets, avoiding subjective differences based on human experience. By utilizing the correlation between data dimensions, indicators with potential impact are mined, improving the comprehensiveness of the evaluation. On the other hand, for the actual monitoring data of the target transmitter, personalized evaluation indicators specific to the target transmitter are mined, improving the targeting of the evaluation. This invention combines historical data mining with the analytic hierarchy process (AHP) to dynamically generate evaluation indicators and weights, improving the accuracy of determining the performance of broadcast transmitters.

[0169] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. When the processor 40 executes the computer program 42, it implements the steps in the various method embodiments described above. Alternatively, when the processor 40 executes the computer program 42, it implements the functions of each module / unit in the various device embodiments described above.

[0170] For example, computer program 42 may be divided into one or more modules / units, which are stored in memory 41 and executed by processor 40 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 42 in electronic device 4.

[0171] Electronic device 4 may include, but is not limited to, processor 40 and memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0172] The processor 40 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0173] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard disk or RAM. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 41 can include both internal and external storage units of the electronic device 4. The memory 41 is used to store the computer program 42 and other programs and data required by the electronic device 4. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0174] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0175] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0176] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0177] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0178] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0179] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for determining the health status and mission reachability of a broadcast transmitter, characterized in that, include: Acquire historical monitoring data of the target transmitter; The historical monitoring data contains multiple data dimensions; Based on the historical monitoring data, frequent itemsets are obtained according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension. Based on the historical monitoring data, frequent itemsets are obtained according to the probability of occurrence of each data dimension combination item at different times. This includes: for any data dimension in the historical monitoring data, the value of the data dimension is converted to mean value and mapped to a trend change; the trend change includes rising, falling, or remaining flat; the trend changes of each data dimension at the same time are treated as a transaction, constructing a transaction set; based on the transaction set, the number of times the combination item of each data dimension's trend change occurs in different transactions is divided by the total number of transactions to obtain the probability of occurrence of the combination item; and combinations with a probability of occurrence greater than a preset value are designated as frequent itemsets. Calculating the association probability between any two data dimensions in the frequent itemset, and selecting multiple principal data dimensions based on the association probability between any two data dimensions in the frequent itemset; calculating the association probability between any two data dimensions in the frequent itemset, and selecting multiple principal data dimensions based on the association probability between any two data dimensions in the frequent itemset, includes: determining the number of times the trend change of any two data dimensions in the frequent itemset occurs in the same transaction; using the ratio of the number of times the trend change of the two data dimensions occurs in the same transaction to the total number of transactions as the association probability of the two data dimensions; if the association probability of the two data dimensions is greater than a preset value, then the two data dimensions are selected as principal data dimensions. We selected data dimensions related to transmitter performance from each master data dimension as evaluation metrics; For each of the aforementioned evaluation indicators, the weight of each evaluation indicator is determined based on the analytic hierarchy process (AHP). The performance of the target transmitter is determined by scoring based on each evaluation indicator and its weight; the transmitter performance includes health or mission reachability.

2. The method for determining the health status and mission reachability of a broadcast transmitter according to claim 1, characterized in that, For any data dimension in historical monitoring data, the values ​​of that data dimension are transformed by mean conversion to map trend changes, including: For any data dimension in historical monitoring data The mean is calculated using the following formula: The following formula maps data points of the same data dimension to the mean to represent trend changes: in, This represents the i-th data point in the data dimension; This represents the mean; This represents the tolerance threshold, indicating the data at... Fluctuations within a certain range are normal.

3. The method for determining the health status and mission reachability of a broadcast transmitter according to claim 1, characterized in that, The transmitter includes multiple power amplifier modules; After filtering to obtain multiple master data dimensions, the following is also included: Each main data dimension of the power amplifier module is refined to obtain the refined main data dimensions; the refinement of each main data dimension of the power amplifier module includes: For each master data dimension of each power amplifier module, the horizontal differences between multiple modules are determined based on the differences between different power amplifier modules; For each master data dimension of each power amplifier module, the fluctuation distance of a single module is determined based on the distance of the master data dimension from the reasonable value range; For each main data dimension of each power amplifier module, the attenuation of a single module is determined based on the attenuation level of the main data dimension; For each master data dimension of each power amplifier module, anomalies in single module data are determined based on whether the master data dimension exceeds a fault threshold.

4. The method for determining the health status and mission reachability of a broadcast transmitter according to claim 1, characterized in that, The transmitter performance refers to the equipment health status; the selection of data dimensions related to transmitter performance from each master data dimension as evaluation indicators includes: Power amplifier temperature, exciter temperature, power amplifier current, exciter current, and overall machine current are used as indicators for assessing the health of the equipment.

5. The method for determining the health status and mission reachability of a broadcast transmitter according to claim 1, characterized in that, The transmitter performance refers to mission reachability. The process of selecting data dimensions related to transmitter performance from each master data dimension as evaluation metrics includes: The power amplifier output power, exciter output power, power amplifier reflected power, and exciter reflected power are used as evaluation indicators for the achievability of equipment tasks.

6. A device for determining the health status and mission reachability of a broadcast transmitter, characterized in that, include: The acquisition module is used to acquire historical monitoring data of the target transmitter; The historical monitoring data contains multiple data dimensions; The frequent itemset acquisition module is used to obtain frequent itemsets based on the historical monitoring data and according to the probability of occurrence of each data dimension combination item at different times; each frequent item includes at least one data dimension. Based on the historical monitoring data, frequent itemsets are obtained according to the probability of occurrence of each data dimension combination item at different times. This includes: for any data dimension in the historical monitoring data, the value of the data dimension is converted to mean value and mapped to a trend change; the trend change includes rising, falling, or remaining flat; the trend changes of each data dimension at the same time are treated as a transaction, constructing a transaction set; based on the transaction set, the number of times the combination item of each data dimension's trend change occurs in different transactions is divided by the total number of transactions to obtain the probability of occurrence of the combination item; and combinations with a probability of occurrence greater than a preset value are designated as frequent itemsets. The main dimension filtering module is used to calculate the association probability between any two data dimensions in the frequent itemset, and to filter out multiple main data dimensions based on the association probability between any two data dimensions in the frequent itemset. Calculating the association probability between any two data dimensions in the frequent itemset and filtering out multiple main data dimensions based on the association probability between any two data dimensions in the frequent itemset includes: determining the number of times the trend changes of any two data dimensions in the frequent itemset occur in the same transaction; using the ratio of the number of times the trend changes of the two data dimensions occur in the same transaction to the total number of transactions as the association probability of the two data dimensions; if the association probability of the two data dimensions is greater than a preset value, then the two data dimensions are selected as main data dimensions. The evaluation metric filtering module is used to filter data dimensions related to transmitter performance from each master data dimension as evaluation metrics. The weight determination module is used to determine the weight of each evaluation indicator based on the analytic hierarchy process. The performance determination module is used to score based on various evaluation indicators and their weights to determine the performance of the target transmitter; the transmitter performance includes health or mission reachability.

7. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method for determining the health status and mission reachability of a broadcast transmitter as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for determining the health status and mission reachability of a broadcast transmitter as described in any one of claims 1 to 5.

Images