Bus alarm controller with low false alarm rate

Abstract

The invention discloses a bus alarm controller with a low false alarm rate, and relates to the field of alarm communication, and the controller comprises a bus alarm module which is used for abutting a plurality of target nodes, serves as a bus end for editing, transmitting and receiving operation instructions of application modules, and carries out the start-stop control of target node function equipment and the application modules, carrying a local and cloud data storage library; the end point acquisition module is used as an acquisition sub-end point deployed at each target node and is used for acquiring operation setting data and abnormal signal data of the alarm feedback target node and the adjacent sub-nodes in real time; according to the method, the future fluctuation performance of the target node and the adjacent nodes can be more accurately predicted, and the alarm level is dynamically adjusted according to the deviation between the abnormal signal acquired in real time and the fluctuation prediction data, so that the alarm level is more flexible and accurate, and can adapt to different operating environments and states, and unnecessary and incorrect alarms are further reduced.

Description

Technical Field

[0001] The invention relates to the technical field of alarm communication, in particular to a bus alarm controller with low false alarm rate. Background Art

[0002] The bus alarm controller is a device used in alarm systems. It is usually used to centrally monitor and control multiple detectors and alarm devices. It can receive signals from various detectors and monitor environmental changes in real time. When a fire or abnormal situation is detected, the bus alarm controller can trigger an alarm, sound and light alarms, and send alarm information to relevant personnel. It also supports linkage control with sprinkler systems, smoke exhaust systems and other equipment. It is configured and managed through software to monitor and maintain the entire alarm system and realize automated emergency response.

[0003] However, traditional alarm control systems are often based on simple threshold judgments. When the monitored signal exceeds a preset threshold, an alarm is triggered. This method is prone to false alarms due to short-term fluctuations or noise, and cannot adjust the alarm threshold in real time. It cannot adapt to changes in system load, operating environment, etc., and only relies on uploaded data from the alarm target. It lacks analysis of the coefficient of change of node parameters in adjacent areas, which makes the system less flexible in various situations and easily leads to false alarms. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a bus alarm controller with a low false alarm rate, which can effectively solve the problems of the prior art.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] The invention discloses a bus alarm controller with low false alarm rate, comprising:

[0009] The bus alarm module is used to connect to several target nodes and serve as the bus terminal for editing, sending and receiving operation instructions of each application module. It controls the start and stop of each target node functional device and each application module, and carries local and cloud data storage libraries.

[0010] The endpoint collection module, as a collection sub-endpoint deployed at each target node, collects the operation setting data and abnormal signal data of the target node with alarm feedback and its adjacent sub-nodes in real time;

[0011] The preprocessing module is used to remove missing values ​​and outliers from the collected data and perform normalization processing;

[0012] The model building module is used to build a fluctuation prediction model, use historical data to train the model, and perform fluctuation prediction based on the target node operation setting data and the collected data of adjacent nodes processed by the preprocessing module;

[0013] A data extraction module, used for obtaining two types of fluctuation performance prediction data of the target node and its adjacent nodes from the fluctuation prediction model output;

[0014] A comprehensive analysis module is used to perform weighted aggregation of two types of volatility performance forecast data according to preset weights and output comprehensive interval volatility performance data;

[0015] The abnormal signal analysis module is used to calculate the difference between the current abnormal signal and the comprehensive fluctuation prediction data to determine whether it is within the safety range;

[0016] The level adjustment module is used to dynamically adjust the alarm level in real time according to the deviation value between the abnormal signal and the fluctuation prediction data.

[0017] Furthermore, the endpoint acquisition module is interactively connected to a data verification module via a wireless network. The data verification module is used to obtain real performance data within a preset period, compare the real performance data with the prediction results, evaluate the accuracy of the model, select the mean square error MSE to quantify the error between the predicted value and the actual value, use optimization algorithms such as gradient descent to minimize the loss function, and output the accuracy coefficient to the model construction module for feedback correction.

[0018] Furthermore, the feedback correction process includes: obtaining the accuracy coefficient to identify the source of error generated in the prediction process, adjusting the hyperparameters of the fluctuation prediction model through a random search algorithm, obtaining features above a preset impact threshold based on the analysis results, removing and adding features, re-encoding the processed features, and retraining the fluctuation prediction model.

[0019] Furthermore, the construction and operation process of the fluctuation prediction model in the model construction module is as follows:

[0020] A regression model algorithm is introduced to obtain the historical operation setting data of the target node and the historical operation data of its adjacent nodes;

[0021] The historical operation setting data of the target node obtained is merged with the historical operation data of the adjacent nodes to form a new data set, and the specific setting parameter characteristics of the current node, the state characteristics and time characteristics of the adjacent nodes are selected to perform data standardization;

[0022] The data is divided into 70% training set and 30% test set, and the regression model is trained using the training set data;

[0023] Use the trained model to predict fluctuations. Input the real-time operation setting data and the latest collected data of adjacent nodes into the model. The model calculates the predicted value of future fluctuations based on the input features to form fluctuation performance prediction data.

[0024] Furthermore, the calculation formula of the regression model algorithm is:

[0025]

[0026] In the formula, y represents the target variable to be predicted, P O represents the intercept term of the model, P j is the regression coefficient of the jth feature in a feature, X j represents the jth feature, E k is the regression coefficient of the kth feature in the adjacent node data, Z k represents the feature data of the kth adjacent node, and W represents the error term.

[0027] Furthermore, the data extraction module obtains two types of fluctuation performance prediction data of the target node and its adjacent nodes as historical data of the target node in the current period, submits them to the model construction module as raw data to be used for training, and performs automatic coverage of the preset period.

[0028] Furthermore, the comprehensive analysis module is interactively connected to a weight adaptive module via a wireless network, and the weight adaptive module is used to actively identify the performance score of each sub-endpoint acquisition device, and analyze based on the current cycle performance score and model prediction performance, and automatically assign weights to each sub-endpoint.

[0029] Furthermore, in the process of automatic allocation, adjustment levels of positive and negative proportional trends are preset, and after the analysis is completed, the weights of the associated sub-endpoints in the next cycle are automatically adjusted according to the corresponding adjustment levels.

[0030] Furthermore, the dynamic adjustment process of the level adjustment module is:

[0031] Set the upper and lower limits of the safety range based on historical adjustment data and custom adjustments;

[0032] If the deviation value is higher than the safety interval, the analysis will adjust the level according to the preset difference and increase the alarm level accordingly. If the deviation value is lower than the safety interval, the analysis will adjust the level according to the preset difference and reduce the alarm level accordingly. If the deviation value is within the safety interval, no change will be made.

[0033] Furthermore, the bus alarm module is interactively connected to the endpoint acquisition module through a wireless network, the level adjustment module is connected to the bus alarm module through an electrical medium, the endpoint acquisition module is interactively connected to the preprocessing module through a wireless network, the preprocessing module is interactively connected to the model building module through a wireless network, the model building module is interactively connected to the data extraction module and the data verification module through a wireless network, the data extraction module is interactively connected to the comprehensive analysis module through a wireless network, the comprehensive analysis module is interactively connected to the abnormal signal analysis module through a wireless network, and the abnormal signal analysis module is interactively connected to the level adjustment module through a wireless network.

[0034] (III) Beneficial effects

[0035] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0036] 1. By constructing a fluctuation prediction model, incorporating multiple variables of the alarm node and its associated nodes into the prediction model, and using historical data for training, the future fluctuation performance of the target node and its adjacent nodes can be predicted more accurately. The alarm level can be dynamically adjusted according to the deviation between the abnormal signal collected in real time and the fluctuation prediction data, making the alarm level more flexible and accurate, and able to adapt to different operating environments and states, thereby reducing unnecessary and incorrect alarms.

[0037] 2. By introducing a feedback mechanism and analyzing the sources of errors in the prediction process, the model parameters are automatically adjusted to ensure the real-time nature of alarm control. By comprehensively using a variety of data processing technologies, fluctuation prediction, dynamic adjustment of alarm levels and feature optimization, the false alarm rate of the system is reduced overall, the reliability of the alarm is improved, and the waste of resources and safety hazards caused by false alarms are effectively prevented.

[0038] 3. By analyzing the performance scores of each sub-endpoint acquisition device and automatically assigning weights based on the model's predicted performance, the control system can adapt to different operating conditions. During long-term monitoring, the deviation of the predicted data is reduced, thereby enhancing the overall stability and responsiveness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 It is a schematic diagram of the framework of the present invention.

[0041] The numbers in the figure represent, respectively, 1. bus alarm module; 2. endpoint acquisition module; 3. preprocessing module; 4. model building module; 5. data extraction module; 6. comprehensive analysis module; 61. weight adaptive module; 7. abnormal signal analysis module; 8. level adjustment module; 9. data verification module. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0043] The present invention will be further described below in conjunction with the embodiments.

[0044] Example 1

[0045] A bus alarm controller with low false alarm rate in this embodiment, such as Figure 1 As shown, including:

[0046] The bus alarm module (1) is used to connect to a number of target nodes, serves as a bus terminal for editing, sending and receiving operation instructions of each application module, performs start and stop control of each target node functional device and each application module, and carries local and cloud data storage libraries; as a centralized control platform, it can monitor and manage all target nodes in real time, ensure efficient communication and coordination between various application modules, support dual storage of local and cloud data, and help provide data backup and realize remote management;

[0047] The endpoint collection module (2), as a collection sub-endpoint deployed at each target node, collects in real time the operation setting data and abnormal signal data of the target node with alarm feedback and its adjacent sub-nodes; thereby ensuring that problems are discovered in time and corresponding measures are taken, and the data integration of adjacent nodes improves the overall monitoring effect;

[0048] The endpoint acquisition module 2 is interactively connected to the data verification module 9 through a wireless network. The data verification module 9 is used to obtain real performance data within a preset period, compare the real performance data with the prediction results, evaluate the accuracy of the model, select the mean square error MSE to quantify the error between the predicted value and the actual value, use optimization algorithms such as gradient descent to minimize the loss function, and output the accuracy coefficient to the model construction module 4 for feedback correction. The feedback correction process includes: obtaining the accuracy coefficient to identify the source of the error generated in the prediction process, adjusting the hyperparameters of the fluctuation prediction model through a random search algorithm, obtaining features above the preset impact threshold according to the analysis results, performing feature elimination and feature addition, re-encoding the processed features, and retraining the fluctuation prediction model;

[0049] Preprocessing module 3, used to remove missing values ​​and abnormal values ​​from the collected data and perform normalization processing;

[0050] Model building module 4 is used to build a fluctuation prediction model, use historical data to train the model, and make fluctuation predictions based on the target node operation setting data and the collected data of adjacent nodes processed by preprocessing module 3; historical data is introduced for training to enhance the prediction ability of the model, so that it can more accurately identify potential abnormal fluctuations, and adjust the model parameters in time according to the actual situation to achieve more accurate early warning;

[0051] A data extraction module 5 is used to obtain two types of fluctuation performance prediction data of the target node and its adjacent nodes from the fluctuation prediction model output;

[0052] The data extraction module 5 obtains two types of fluctuation performance prediction data of the target node and its adjacent nodes as the historical data of the target node in the current period, submits them to the model construction module 4 as the original data to be used for training, and performs automatic coverage of the preset period;

[0053] Comprehensive analysis module 6 is used to perform weighted aggregation of two types of fluctuation performance prediction data according to preset weights and output comprehensive interval fluctuation performance data; improve the accuracy of fluctuation performance prediction, make the decision basis more comprehensive, and reduce the probability of failure;

[0054] The abnormal signal analysis module 7 is used to calculate the difference between the current abnormal signal and the comprehensive fluctuation prediction data to determine whether it is within the safe range; it monitors the difference between the current abnormal signal and the prediction data in real time, and promptly determines the safe state of system operation. Through continuous comparison, it can quickly identify abnormal situations and activate the alarm mechanism to help operation and maintenance personnel respond quickly;

[0055] The level adjustment module 8 is used to dynamically adjust the alarm level in real time according to the deviation value between the abnormal signal and the fluctuation prediction data. The dynamic adjustment process of the level adjustment module 8 is as follows:

[0056] Set the upper and lower limits of the safety range based on historical adjustment data and custom adjustments;

[0057] If the deviation value is higher than the safety interval, the alarm level will be increased accordingly according to the preset difference adjustment level. If the deviation value is lower than the safety interval, the alarm level will be decreased accordingly according to the preset difference adjustment level. If the deviation value is within the safety interval, no change will be made. The alarm level will be adjusted intelligently according to real-time data and historical adjustment data to achieve accurate alarm management.

[0058] As a preferred implementation in this embodiment, Figure 1 As shown, the bus alarm module 1 is interactively connected to the endpoint acquisition module 2 through a wireless network, the level adjustment module 8 is connected to the bus alarm module 1 through an electrical medium, the endpoint acquisition module 2 is interactively connected to the preprocessing module 3 through a wireless network, the preprocessing module 3 is interactively connected to the model building module 4 through a wireless network, the model building module 4 is interactively connected to the data extraction module 5 and the data verification module 9 through a wireless network, the data extraction module 5 is interactively connected to the comprehensive analysis module 6 through a wireless network, the comprehensive analysis module 6 is interactively connected to the abnormal signal analysis module 7 through a wireless network, and the abnormal signal analysis module 7 is interactively connected to the level adjustment module 8 through a wireless network.

[0059] This embodiment significantly improves the prediction accuracy and operational safety of the system through multi-level data processing and intelligent algorithms, thereby achieving efficient management and control of target nodes and adapting to complex and changing industrial environments.

[0060] Example 2

[0061] In other aspects, this embodiment also provides a fluctuation prediction model. The construction and operation process of the fluctuation prediction model in the model construction module 4 is as follows:

[0062] A regression model algorithm is introduced to obtain the historical operation setting data of the target node and the historical operation data of its adjacent nodes;

[0063] The historical operation setting data of the target node obtained is merged with the historical operation data of the adjacent nodes to form a new data set, and the specific setting parameter characteristics of the current node, the state characteristics and time characteristics of the adjacent nodes are selected to perform data standardization;

[0064] The data is divided into 70% training set and 30% test set, and the regression model is trained using the training set data;

[0065] Use the trained model to predict fluctuations, input the real-time operation setting data and the latest collected data of adjacent nodes into the model, and the model calculates the predicted value of future fluctuations based on the input features to form fluctuation performance prediction data; use the operation settings of the current node and the historical status information of the adjacent nodes to establish a regression relationship and predict future fluctuation performance, thereby providing data support for alarm control. Reasonable feature selection and model optimization can greatly improve the prediction accuracy and reduce the probability of false alarms.

[0066] The calculation formula of the regression model algorithm is:

[0067]

[0068] In the formula, y represents the target variable for prediction, that is, the predicted value of future fluctuations, P O Represents the intercept term of the model, which indicates the predicted target value when all independent variable characteristics are zero, P j is the regression coefficient of the jth feature in a certain feature, indicating the influence of this feature on the volatility prediction value, X j represents the jth feature, such as a certain operating setting parameter of the target node, E k is the regression coefficient of the kth feature in the adjacent node data, indicating the influence of the adjacent feature on the fluctuation prediction value, Z k represents the characteristic data of the kth adjacent node, and W represents the error term, which reflects the random noise or influencing factors that the model cannot explain.

[0069] Example 3

[0070] In this embodiment, the comprehensive analysis module 6 is interactively connected to the weight adaptive module 61 through a wireless network. The weight adaptive module 61 is used to actively identify the performance score of each sub-endpoint acquisition device, and analyze based on the current cycle performance score and model prediction performance, and automatically assign weights to each sub-endpoint. During the automatic assignment process, adjustment levels of positive and inverse trends are preset. After the analysis is completed, the weights of the associated sub-endpoints in the next cycle are automatically adjusted according to the corresponding adjustment levels.

[0071] Compared with existing technologies, the ability to dynamically adjust weights based on real-time performance scores and model prediction results helps to respond to performance changes of sub-endpoints in a timely manner and ensure the optimal configuration of resources. The preset adjustment levels can refine the handling of different situations. This flexibility can reasonably optimize for different usage scenarios, reduce reliance on manual intervention, and by comprehensively considering the data of multiple sub-endpoints, potential performance bottlenecks and optimization points can be discovered, thereby achieving global performance improvement.

[0072] Working principle: The present invention obtains the control authority of each node device in the current control area through the bus alarm module 1, controls the global function module and function equipment, deploys the endpoint acquisition module 2 to the function equipment of each node, obtains the node monitoring data, pre-processes the collected data through the pre-processing module 3, and constructs the fluctuation prediction model through the model construction module 4. After receiving the alarm signal, the operation data of the target node and its adjacent nodes are called into the model for calculation, and the calculated two types of prediction data are extracted by the data extraction module 5. The two types of prediction data are integrated and analyzed according to the preset weights through the comprehensive analysis module 6, and the final prediction data is output. The weight adaptive module 61 provides a reference for setting the weight by analyzing the actual performance of the model, calculates the deviation value between the current alarm data and the final prediction data through the abnormal signal analysis module 7, edits the adjustment instruction, and dynamically adjusts the alarm level through the level adjustment module 8. The prediction performance of the model is verified by the data verification module 9 with real data;

[0073] The present invention adopts an online learning strategy, allowing the controller to dynamically update the model according to real-time data, which can adapt to the changing operating environment and load conditions in a timely manner, incorporate multiple variables of the alarm node and its associated nodes into the prediction model, and comprehensively capture the system status, which helps to improve the prediction accuracy. Through the feedback mechanism, the controller can adjust the model parameters according to the actual alarm data to reduce the limitations of the source of abnormal data, thereby continuously reducing false alarms. The real-time comparison of the comprehensive interval fluctuation performance data and abnormal signals improves the decision-making efficiency of operation and maintenance personnel and helps them make more scientific and reasonable response measures.

[0074] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will 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.

Claims

1. A bus alarm controller with low false alarm rate, characterized in that: include: The bus alarm module (1) is used to connect to a number of target nodes, serve as a bus terminal for editing, sending and receiving operation instructions of each application module, perform start and stop control of each target node functional device and each application module, and carry local and cloud data storage libraries; The endpoint collection module (2), as a collection sub-endpoint deployed at each target node, collects the operation setting data and abnormal signal data of the target node with alarm feedback and its adjacent sub-nodes in real time; A preprocessing module (3) is used to remove missing values ​​and abnormal values ​​from the collected data and perform normalization processing; A model building module (4) is used to build a fluctuation prediction model, use historical data to train the model, and perform fluctuation prediction based on the target node operation setting data and the collected data of adjacent nodes processed by the preprocessing module (3); A data extraction module (5), used for obtaining two types of fluctuation performance prediction data of the target node and its adjacent nodes from the fluctuation prediction model output; A comprehensive analysis module (6), used for weighting and summarizing the two types of volatility performance prediction data according to preset weights, and outputting comprehensive interval volatility performance data; An abnormal signal analysis module (7), used to calculate the difference between the current abnormal signal and the comprehensive fluctuation prediction data, and determine whether it is within the safety range; The level adjustment module (8) is used to dynamically adjust the alarm level in real time according to the deviation value between the abnormal signal and the fluctuation prediction data.

2. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The endpoint acquisition module (2) is interactively connected to a data verification module (9) via a wireless network. The data verification module (9) is used to obtain real performance data within a preset period, compare the real performance data with the prediction results, evaluate the accuracy of the model, select mean square error (MSE) to quantify the error between the prediction value and the actual value, use optimization algorithms such as gradient descent to minimize the loss function, and output the accuracy coefficient to the model construction module (4) for feedback correction.

3. A bus alarm controller with low false alarm rate according to claim 2, characterized in that: The feedback correction process includes: obtaining the accuracy coefficient to identify the source of error generated in the prediction process, adjusting the hyperparameters of the fluctuation prediction model through a random search algorithm, obtaining features above a preset impact threshold based on the analysis results, removing and adding features, re-encoding the processed features, and retraining the fluctuation prediction model.

4. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The construction and operation process of the fluctuation prediction model in the model construction module (4) is as follows: A regression model algorithm is introduced to obtain the historical operation setting data of the target node and the historical operation data of its adjacent nodes; The historical operation setting data of the target node obtained is merged with the historical operation data of the adjacent nodes to form a new data set, and the specific setting parameter characteristics of the current node, the state characteristics and time characteristics of the adjacent nodes are selected to perform data standardization; The data is divided into 70% training set and 30% test set, and the regression model is trained using the training set data; Use the trained model to predict fluctuations. Input the real-time operation setting data and the latest collected data of adjacent nodes into the model. The model calculates the predicted value of future fluctuations based on the input features to form fluctuation performance prediction data.

5. A bus alarm controller with low false alarm rate according to claim 4, characterized in that: The calculation formula of the regression model algorithm is: In the formula, y represents the target variable for prediction, that is, the predicted value of future fluctuations, P O Represents the intercept term of the model, which indicates the predicted target value when all independent variables (features) are zero, P j is the regression coefficient of the jth feature in a certain feature, indicating the influence of this feature on the volatility prediction value, X j represents the jth feature, such as a certain operating setting parameter of the target node, E k is the regression coefficient of the kth feature in the adjacent node data, indicating the influence of the adjacent feature on the fluctuation prediction value, Z k represents the characteristic data of the kth adjacent node, and W represents the error term, which reflects the random noise or influencing factors that the model cannot explain.

6. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The data extraction module (5) obtains two types of fluctuation performance prediction data of the target node and its adjacent nodes as historical data of the target node in the current period, submits them to the model construction module (4) as raw data to be used for training, and performs automatic coverage of a preset period.

7. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The comprehensive analysis module (6) is interactively connected to a weight adaptive module (61) via a wireless network. The weight adaptive module (61) is used to actively identify the performance score of each sub-endpoint acquisition device, and analyze based on the current cycle performance score and model prediction performance, and automatically allocate the weight of each sub-endpoint.

8. A bus alarm controller with low false alarm rate according to claim 7, characterized in that: During the automatic allocation process, the adjustment levels of the positive and negative trends are preset, and after the analysis is completed, the weights of the associated sub-endpoints in the next cycle are automatically adjusted according to the corresponding adjustment levels.

9. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The dynamic adjustment process of the level adjustment module (8) is as follows: Set the upper and lower limits of the safety range based on historical adjustment data and custom adjustments; If the deviation value is higher than the safety interval, the analysis will adjust the level according to the preset difference and increase the alarm level accordingly. If the deviation value is lower than the safety interval, the analysis will adjust the level according to the preset difference and reduce the alarm level accordingly. If the deviation value is within the safety interval, no change will be made.

10. A bus alarm controller with low false alarm rate according to claim 1, characterized in that: The bus alarm module (1) is interactively connected to the endpoint acquisition module (2) via a wireless network, the level adjustment module (8) is connected to the bus alarm module (1) via an electrical medium, the endpoint acquisition module (2) is interactively connected to the preprocessing module (3) via a wireless network, the preprocessing module (3) is interactively connected to the model construction module (4) via a wireless network, the model construction module (4) is interactively connected to the data extraction module (5) and the data verification module (9) via a wireless network, the data extraction module (5) is interactively connected to the comprehensive analysis module (6) via a wireless network, the comprehensive analysis module (6) is interactively connected to the abnormal signal analysis module (7) via a wireless network, and the abnormal signal analysis module (7) is interactively connected to the level adjustment module (8) via a wireless network.

Images