Intelligent early warning car washer

Abstract

The invention discloses an intelligent early warning car washer, which relates to the technical field of car washers, and comprises a data acquisition module, a data processing and feature extraction module, an anomaly detection module, a fault prediction and early warning module and a false alarm correction and verification module: the data acquisition module acquires operation data of the car washer through a preset sensor and monitoring equipment; according to the invention, through adaptive threshold adjustment, multi-dimensional data fusion and machine learning technologies, the intelligent early warning system significantly improves the accuracy of fault early warning, reduces false alarm and missing alarm, and ensures that an alarm is triggered only when a real fault occurs. The system can adjust the threshold value in real time, integrates a plurality of key parameters, accurately predicts equipment faults, and helps to discover potential risks in time. And through accurate early warning and intelligent maintenance, the downtime of the equipment is reduced, major faults and high maintenance cost are avoided, the service life of the equipment is prolonged, and the working efficiency and the economic benefits of enterprises are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of car wash machines, and in particular to an intelligent early warning car wash machine. Background Art

[0002] The intelligent early warning car washing machine is an automated car washing equipment that combines advanced sensor technology, the Internet of Things (IoT) and data analysis. It is designed to monitor various operating indicators in the car washing process in real time through an intelligent system, automatically identify equipment failures or potential problems, and issue early warning signals in advance. The system can detect the working status of the car washing machine's motor, electronic control system, water pump, nozzle, pipe and other components, such as abnormal fluctuations in parameters such as temperature, pressure, and current. Once a component is found to be about to fail, the system will automatically trigger an alarm to remind maintenance personnel to repair or replace it in time to avoid serious failures of the equipment during operation, thereby effectively improving the operating efficiency of the car washing machine, extending the service life of the equipment, and reducing downtime, ensuring the continuity and reliability of service.

[0003] The prior art has the following deficiencies: The existing intelligent early warning car washing machine may face the problem of false alarm and missed alarm of the early warning system. If the intelligent early warning system cannot accurately judge the operating status of the equipment, false alarms may occur, causing maintenance personnel to conduct frequent and unnecessary inspections or replace parts, thereby increasing operating costs and the frequency of manual intervention. At the same time, missed alarms may cause potential faults to fail to be identified in time, causing equipment failure without repair, and further causing serious equipment damage or downtime. These problems may affect the normal operation of the car washing machine and cause service interruptions, which will seriously affect the user experience and the long-term reliability of the equipment.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent early warning car washing machine. Through adaptive threshold adjustment, multi-dimensional data fusion and machine learning technology, the intelligent early warning system significantly improves the accuracy of fault warning, reduces false alarms and missed alarms, and ensures that the alarm is triggered only when a real fault occurs. The system can adjust the threshold in real time, integrate multiple key parameters and accurately predict equipment failures, helping to discover potential risks in a timely manner. Through accurate early warning and intelligent maintenance, equipment downtime is reduced, major failures and high maintenance costs are avoided, the service life of the equipment is extended, and work efficiency and corporate economic benefits are improved to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent early warning car washing machine, including a data acquisition module, a data processing and feature extraction module, an abnormality detection module, a fault prediction and early warning module, and a false alarm correction and verification module: The data collection module collects the operation data of the car washing machine through preset sensors and monitoring equipment; The data processing and feature extraction module processes the collected data in real time, removes noise through filtering algorithms, and extracts key feature parameters related to the equipment operation status; The anomaly detection module uses an adaptive threshold judgment algorithm to determine whether each key feature parameter exceeds the normal working range based on historical data and set thresholds; The fault prediction and early warning module uses a fault prediction algorithm and a machine learning model to train historical fault data, predict potential fault types and occurrence probabilities, and generate corresponding early warning signals; The false alarm correction and verification module uses a false alarm correction algorithm based on multi-dimensional joint judgment to further verify the warning results, and performs weighted analysis on the abnormal detection results according to the real-time equipment status, environmental conditions and historical fault patterns to reduce the probability of false alarms and missed alarms, and ensure the accuracy and reliability of the warning information.

[0007] Preferably, the specific steps of collecting the operation data of the car washing machine through the preset sensors and monitoring equipment are as follows: Collect key operating data of equipment in real time through sensors; To ensure data accuracy, a timestamp is added to each piece of data so that the order of the data can be maintained during subsequent analysis; Screen the collected data, remove errors and outliers, and ensure data quality; The collected data is stored in a data center or cloud server to ensure the real-time and integrity of data transmission.

[0008] Preferably, the specific steps of processing the collected data in real time, removing noise through a filtering algorithm, and extracting key characteristic parameters related to the operating status of the equipment are as follows: Use filtering algorithms to remove noise caused by external interference to ensure the smoothness and accuracy of data; By eliminating outliers and smoothing data fluctuations, we ensure that data is not affected by instantaneous interference; Extract key characteristic parameters reflecting the health status of the equipment from the processed data; The extracted feature data are standardized so that different feature parameters have the same dimension, which is convenient for subsequent analysis.

[0009] Preferably, the specific steps of using an adaptive threshold judgment algorithm to judge whether each key characteristic parameter exceeds the normal working range according to historical data and set thresholds are as follows: By analyzing historical operation data, a normal working status model of the equipment is established to provide a basis for setting thresholds; Set thresholds based on historical data and adjust them in real time through adaptive algorithms to adapt to different operating conditions; Compare real-time data with set thresholds. When the threshold is exceeded, it is marked as an abnormal state and an alarm is triggered. Abnormal conditions are marked and classified in a timely manner so that different levels of processing measures and feedback can be taken to ensure the normal operation of the equipment.

[0010] Preferably, the specific steps of training historical fault data through a fault prediction algorithm in combination with a machine learning model, predicting potential fault types and occurrence probabilities, and generating corresponding warning signals are as follows: Before fault prediction, the historical fault data is preprocessed and feature engineered to provide input data for the machine learning model. By analyzing the operating status of the equipment before and after the fault, the factors that potentially affect the occurrence of the fault are extracted. The formula is as follows: , where is the original fault data matrix, is the processed data matrix, It is a data standardization operation. is a feature extraction operation; Using processed data The machine learning model is trained with labeled data. The model learns the characteristic performance of the device in normal and faulty states to predict the potential fault types and their probability of occurrence in the future. The formula is as follows , where is the optimal model parameter set obtained by minimizing the mean square error loss function, The model predicts The failure probability of a sample, It is the first in the historical fault data The true fault labels of samples, is the number of samples, is to find the parameters that minimize the loss function ; The model obtained through training , predict the newly collected real-time data to obtain the probability of equipment failure. The probability value will provide a basis for the subsequent early warning signal generation. The probability value calculation formula is: , where is the predicted probability of failure occurrence, is the newly collected real-time device data matrix, is the bias term, is the transpose of the matrix, yes Activation function; According to the probability of failure , according to the set threshold, it is determined whether to generate an early warning signal. The early warning signal provides a basis for subsequent decision-making. The judgment formula is as follows: , where It is a warning signal. If the predicted probability is greater than or equal to the threshold, it is 1 to trigger a warning, otherwise it is 0 and no warning is triggered. is the set fault occurrence probability threshold.

[0011] Preferably, a false alarm correction algorithm based on multi-dimensional joint judgment is used to further verify the warning results, and a weighted analysis is performed on the abnormal detection results according to the real-time equipment status, environmental conditions and historical fault rules to reduce the probability of false alarms and missed alarms. The specific steps to ensure the accuracy and reliability of the warning information are as follows: First, the real-time collected equipment data is scored in multiple dimensions to evaluate the current operating status of the equipment. The specific evaluation formula is as follows: , where is the comprehensive status score of the device. is the number of features involved in scoring, It is The weight of each feature in the comprehensive score, It is The real-time collection value of characteristic parameters, It is Normal working threshold of each feature; In order to consider the impact of the external environment, environmental factors are introduced to make weighted adjustments to the equipment status score. The calculation formula for the adjustment factor of environmental conditions is as follows: , where is the weighted adjustment factor for environmental conditions, is the current environmental condition, is the standard normal environmental condition for the equipment, is the adjustment factor, is the nonlinear adjustment index; The abnormal detection results are corrected in combination with historical fault data. The correction factor calculation formula of historical fault mode is as follows: , where is the correction factor for the historical failure mode, It is The weight coefficient of the failure mode, It is The historical probability of occurrence of a failure mode, It is Equipment status score under various failure modes, is the total number of failure modes; Finally, combined with the comprehensive status score of the equipment , weighted adjustment factors for environmental conditions and correction factors for historical failure modes , comprehensively correct the warning results and obtain the final abnormal judgment probability. The calculation formula is as follows: , where is the corrected abnormality judgment probability, is the adjustment factor, is the natural base, is the historical probability of occurrence of the failure mode.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are: Through adaptive threshold adjustment, multi-dimensional data fusion and machine learning technology, the intelligent early warning system of the present invention can greatly improve the accuracy of fault warnings and avoid the problems of false alarms and missed alarms caused by traditional static threshold judgments. The system dynamically adjusts the threshold according to the real-time working status and environmental conditions of the equipment, and comprehensively considers multiple key parameters to ensure that the early warning signal is only triggered when a real fault occurs, reducing unnecessary alarms and more accurately identifying potential fault risks. In addition, the continuous learning and optimization of historical data by the machine learning model enables the system to accurately predict the timing and type of equipment failures, further improving the reliability and accuracy of the early warning.

[0013] Through accurate fault warning and intelligent maintenance, the system of the present invention can detect potential equipment failure risks in advance, thereby avoiding downtime and high maintenance costs caused by major equipment failures. Through early warning, fault prediction and real-time monitoring, the system helps maintenance personnel take preventive measures in a timely manner, effectively reduces the occurrence of sudden equipment failures, and ensures the continuous and stable operation of the equipment. This not only significantly reduces equipment downtime and improves work efficiency, but also avoids equipment being in a high load or fault state for a long time, slows down the aging of the equipment, thereby extending the service life of the equipment and improving the overall economic benefits of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 The present invention is a module schematic diagram of an intelligent early warning car washing machine. DETAILED DESCRIPTION

[0016] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0017] The present invention provides Figure 1 The intelligent early warning car washing machine shown includes a data acquisition module, a data processing and feature extraction module, an anomaly detection module, a fault prediction and early warning module, and a false alarm correction and verification module: The data acquisition module collects the operation data of the car washing machine through preset sensors and monitoring equipment. The data includes but is not limited to physical parameters such as current, voltage, temperature, pressure, and operation time; The specific steps for collecting the operation data of the car washing machine through the preset sensors and monitoring equipment, including but not limited to current, voltage, temperature, pressure, operating time and other physical parameters, are as follows: Collect key operating data of equipment in real time through sensors, such as current, voltage, temperature, pressure, etc.

[0018] During the operation of the car wash machine, the preset sensors and monitoring equipment continuously collect the physical parameter data of each key component of the equipment, such as current, voltage, temperature, pressure, running time, etc. These parameters provide a comprehensive view of the equipment's operating status and can reflect the working conditions of each component. The data is collected in real time and at a high frequency to capture any abnormalities that may occur in the equipment in a timely manner. This data is transmitted to the central control system or data processing module via wireless or wired means.

[0019] To ensure data accuracy, a timestamp is added to each piece of data so that the order of the data can be maintained during subsequent analysis; All collected data will be timestamped to ensure that the data can be sorted in the correct time sequence. Timestamps are the basis of data synchronization, which helps the system accurately capture instantaneous changes in different device states. For example, under high temperature or high pressure, the current or voltage of a car washing machine may fluctuate, and the timestamp ensures that the data can be correctly associated with the specific device state. This process lays the foundation for subsequent data processing and analysis.

[0020] Screen the collected data, remove errors and outliers, and ensure data quality; During data collection, a preliminary quality check of the collected data is required. This process will screen out obviously unreasonable and abnormal data values ​​(such as instantaneous spikes or noise data), which may be caused by external interference or sensor failure. Through a certain quality control mechanism, it is ensured that only valid and real operating data is sent to the next step of data processing. This step helps to improve the reliability of the data and reduce the impact of noise data on subsequent analysis.

[0021] Store the collected data in a data center or cloud server to ensure the real-time and integrity of data transmission; All collected real-time data will be stored in the data center or cloud server for subsequent processing and analysis. The data transmission process must ensure real-time and integrity, and use a stable communication protocol to avoid data loss or delay. The key to this step is to ensure that the data will not be affected by network fluctuations during transmission, so as to ensure that subsequent processing can proceed smoothly.

[0022] The data processing and feature extraction module processes the collected data in real time, removes noise through filtering algorithms, and extracts key feature parameters related to the equipment operation status; The specific steps for real-time processing of the collected data, removing noise through filtering algorithms, and extracting key characteristic parameters related to the equipment operating status are as follows: Use filtering algorithms to remove noise caused by external interference to ensure the smoothness and accuracy of data; After obtaining the raw data, the data must first be processed by noise filtering. Since the signals collected by the sensor are often affected by a variety of external factors (such as electromagnetic interference, mechanical vibration of the equipment, changes in ambient temperature, etc.), the data may contain some irrelevant noise. These noises may cause inaccurate data, which in turn affects the results of subsequent analysis. Therefore, appropriate filtering algorithms (such as Kalman filtering, mean filtering, wavelet transform, etc.) are used to remove these irrelevant noises to ensure the smoothness and reliability of the data.

[0023] By eliminating outliers and smoothing data fluctuations, we ensure that data is not affected by instantaneous interference; The core purpose of the filtering process is to remove abnormal fluctuations in the data. By setting a certain tolerance threshold, the filtering algorithm can identify and remove parts of the data that do not conform to the normal fluctuation pattern, especially drastic fluctuations in a short period of time (such as instantaneous changes caused by the external environment). For example, abnormal fluctuations in the current of a car washing machine may only be caused by instantaneous voltage changes. If such data is not removed, it will affect the accuracy of subsequent analysis of the system. By smoothing the data, the filtering algorithm effectively reduces these interference factors.

[0024] Extract key characteristic parameters reflecting the health status of the equipment from the processed data; From the filtered data, key characteristic parameters related to the equipment's operating status are extracted. These characteristic parameters are usually important indicators of the equipment's health status. For example, the fluctuation amplitude of the current can reflect changes in the motor load, and changes in temperature may indicate the risk of equipment overheating. By calculating the mean, variance, maximum value, minimum value and other statistics of these characteristic parameters, the core information reflecting the equipment's performance can be extracted from the overall data. The data after feature extraction is more concise and highly representative, providing a solid foundation for subsequent anomaly detection, fault prediction, etc.

[0025] Standardize the extracted feature data so that different feature parameters have the same dimension, which is convenient for subsequent analysis; After feature extraction, it is usually necessary to perform standardization or normalization. This is to prevent the dimensional differences between different feature parameters from affecting the subsequent analysis results. By standardizing each feature parameter (such as converting the data into a normal distribution with a mean of 0 and a standard deviation of 1), each feature can be within the same scale range, so that subsequent machine learning algorithms or anomaly detection algorithms can perform analysis more effectively. This process ensures data consistency, allowing the analysis algorithm to predict the health status of the equipment with higher accuracy.

[0026] The anomaly detection module uses an adaptive threshold judgment algorithm to determine whether each key feature parameter exceeds the normal working range based on historical data and the set threshold. If the parameter exceeds the set threshold, it is marked as an abnormal state; Based on historical data and set thresholds, an adaptive threshold judgment algorithm is used to determine whether each key feature parameter exceeds the normal working range. If the parameter exceeds the set threshold, it is marked as an abnormal state. The specific steps are as follows: By analyzing historical operation data, a normal working status model of the equipment is established to provide a basis for setting thresholds; First, it is necessary to use historical operation data to model the normal working state of the equipment. Historical data includes normal operating parameters, fault records and environmental factors of the equipment. The system establishes a model of normal working state by analyzing these data. These data will provide a basis for setting the initial threshold range. For example, when the equipment is working normally, the current, temperature, pressure and other parameters of the motor have a certain fluctuation range. By statistically analyzing the historical data, the system can derive a rough "normal working range" and deviation range, which can provide a reference for the subsequent adaptive threshold judgment algorithm.

[0027] Set thresholds based on historical data and adjust them in real time through adaptive algorithms to adapt to different operating conditions; Based on the results of historical data analysis, the system sets preliminary thresholds for each key characteristic parameter. These thresholds are usually determined according to the manufacturing standards or industry standards of the equipment. However, since the equipment will age over time or exhibit different working conditions in different operating environments, static thresholds may not meet the working requirements in all situations. Therefore, the system also needs to adopt an adaptive threshold judgment algorithm to adjust these thresholds in real time during the actual operation of the equipment. The adaptive threshold will be dynamically updated according to the real-time collected data. For example, when the temperature gradually rises within a certain period of time, the system will increase the temperature threshold accordingly to avoid false alarms. This dynamic adjustment mechanism enables the system to adjust the threshold according to the current operating status of the equipment and environmental factors, ensuring the flexibility and accuracy of the judgment.

[0028] Compare real-time data with set thresholds. When the threshold is exceeded, it is marked as an abnormal state and an alarm is triggered. Once filtering and feature extraction are completed, the system will compare the real-time data with the set adaptive threshold. When a key characteristic parameter exceeds the set threshold, the system will immediately identify it as an abnormal state and trigger an alarm. For example, if the current of the motor exceeds the normal operating range, or the temperature rises suddenly, the system will immediately mark it as an "abnormal" state. It is worth noting that the system's abnormal judgment does not only rely on the exceeding of a single parameter, but also comprehensively considers multi-dimensional data. For example, if the current and temperature are abnormal at the same time, the system may judge it as a more serious fault risk. At this time, the system will prompt the operator to take further inspection or repair measures through an alarm or early warning.

[0029] Abnormal conditions are marked and classified in a timely manner so that different levels of processing measures and feedback can be taken to ensure the normal operation of the equipment; When a parameter exceeds the normal range, the system will mark it as an abnormal state and record it for subsequent analysis. At the same time, the system will classify the abnormal state to determine whether it is a minor deviation, a warning level, or a major failure. For example, an increase in motor temperature may only be a minor deviation due to increased workload, but a rapid increase in temperature accompanied by an increase in current may indicate a device failure. The system will give different processing suggestions based on the type of abnormality, and provide real-time feedback through the background management platform or the maintenance personnel's operation interface to help maintenance personnel take timely repairs or further monitoring measures to ensure that the equipment returns to normal operation in the shortest time.

[0030] The fault prediction and early warning module uses a fault prediction algorithm and a machine learning model to train historical fault data, predict potential fault types and occurrence probabilities, and generate corresponding early warning signals; The specific steps of using the fault prediction algorithm and the machine learning model to train historical fault data, predict potential fault types and occurrence probabilities, and generate corresponding warning signals are as follows: Before fault prediction, the historical fault data is first preprocessed and feature engineered to provide input data for the machine learning model. Data preprocessing includes cleaning missing values, standardization and normalization operations, while feature engineering involves extracting representative and predictive features from the original data. By analyzing the operating status of the equipment before and after the fault, the factors that potentially affect the occurrence of the fault, such as current fluctuations, temperature changes, pressure changes, etc., are extracted. The formula is as follows: , where is the original fault data matrix, including the operating parameters of the equipment (such as current, temperature, pressure, etc.), is the processed data matrix, the data after normalization and feature extraction, It is a data standardization operation to ensure that the feature parameters have the same dimension, which is convenient for machine learning model processing. It is a feature extraction operation, which extracts features related to the occurrence of faults from historical data; It will serve as input data for subsequent machine learning model training and fault prediction.

[0031] Using processed data The machine learning model is trained with label data (fault type or occurrence status label). The model learns the characteristic performance of the device in normal and faulty states to predict the potential fault type and its probability of occurrence in the future. The formula is as follows , where is the optimal model parameter set obtained by minimizing the mean square error (MSE) loss function, The model predicts The failure probability of a sample, It is the first in the historical fault data The true fault label of samples (0 for normal, 1 for fault), is the number of samples, is to find the parameters that minimize the loss function ; The model obtained through training , predict the newly collected real-time data to obtain the probability of equipment failure. This probability reflects the possibility of a certain failure of the equipment, and risk assessment can be performed based on the prediction results. The probability value will provide a basis for the subsequent generation of early warning signals. The probability value calculation formula is: , where is the predicted probability of failure occurrence, It is a newly collected real-time device data matrix, including current, temperature, pressure and other parameters. is the bias term, is the transpose of the matrix, yes An activation function is used to convert the predicted value into a probability value between 0 and 1, i.e. the probability of a device failure. ; According to the probability of failure , determine whether to generate an early warning signal based on the set threshold. If If the set threshold is exceeded, the system will trigger an early warning and remind maintenance personnel to take timely measures to prevent the occurrence of failures. The early warning signal provides a basis for subsequent decision-making, such as inspection, repair or replacement of equipment. The judgment formula is as follows: , where It is a warning signal. If the predicted probability is greater than or equal to the threshold, it is 1 to trigger a warning, otherwise it is 0 and no warning is triggered. is the set fault occurrence probability threshold.

[0032] The false alarm correction and verification module uses a false alarm correction algorithm based on multi-dimensional joint judgment to further verify the warning results. It performs weighted analysis on abnormal detection results according to real-time equipment status, environmental conditions and historical fault patterns to reduce the probability of false alarms and missed alarms, and ensure the accuracy and reliability of warning information. The false alarm correction algorithm based on multi-dimensional joint judgment is used to further verify the warning results. The abnormal detection results are weighted and analyzed according to the real-time equipment status, environmental conditions and historical fault patterns to reduce the probability of false alarms and missed alarms. The specific steps to ensure the accuracy and reliability of the warning information are as follows: First, the real-time collected equipment data is scored in multiple dimensions to evaluate the current operating status of the equipment. The equipment status score is obtained by weighted calculation of multiple key characteristic parameters (such as current, voltage, temperature, etc.) and set thresholds. The specific evaluation formula is as follows: , where It is the comprehensive status score of the device, which measures the deviation degree of the current status of the device. is the number of features involved in scoring, such as current, voltage, temperature, etc. It is The weight of each feature in the comprehensive score is usually set according to its impact on the operation of the device. It is Real-time collection values ​​of characteristic parameters (such as current, voltage, temperature, etc.), It is Normal working threshold of each feature; Through this step, the deviation between each feature and the normal working threshold can be calculated. The larger the deviation, the more abnormal the device status. The higher it is, the further the device is from normal working state.

[0033] In order to consider the impact of the external environment, environmental factors (such as temperature, humidity, etc.) are introduced to make weighted adjustments to the equipment status score. Changes in environmental conditions may affect the working status of the equipment, so it is necessary to correct the equipment score. The calculation formula for the adjustment factor of environmental conditions is as follows: , where is a weighted adjustment factor for environmental conditions, used to correct the equipment status score, are the current environmental conditions (such as the current temperature or humidity), is the standard normal environmental condition for the equipment, It is the adjustment coefficient, which represents the linear adjustment factor of the impact of environmental conditions on the equipment status. It is usually used to control the basic weight of environmental impact. is a nonlinear adjustment index used to adjust for current environmental conditions and normal environmental conditions of the equipment standards The nonlinear influence factor is often set to a value greater than 1 to amplify the degree of deviation of environmental conditions; This step can correct the device status score by weighting the current environmental conditions. Especially in harsh environments (such as high temperature or high humidity), the normal working state of the device will be affected, so correction is needed to prevent false alarms caused by environmental changes.

[0034] The abnormal detection results are corrected in combination with historical fault data. Historical fault data can reveal the performance of the equipment under different fault modes. Therefore, incorporating this information into the current analysis can further improve the accuracy of the early warning. The correction factor calculation formula for the historical fault mode is as follows: , where is the correction factor for the historical failure mode, It is The weight coefficient of a fault mode indicates the importance of this mode to abnormal correction. It is The historical probability of occurrence of a failure mode, It is The device status score under a certain fault mode indicates the correlation between the current device status and the fault mode. is the total number of failure modes; Through historical fault data, the equipment status score can be weighted and adjusted according to the probability of each fault mode. If a certain fault mode has occurred frequently and has a large impact in history, then the score of this fault mode will have a greater impact on the calculation of the correction factor, thereby more accurately reflecting the health status of the equipment.

[0035] Finally, combined with the comprehensive status score of the equipment , weighted adjustment factors for environmental conditions and correction factors for historical failure modes , comprehensively correct the warning results and obtain the final abnormal judgment probability. The calculation formula is as follows: , where is the corrected abnormality judgment probability, indicating the possibility of equipment failure. is the adjustment factor used to control the effect of the correction on the final result, is the natural base, is the historical probability of occurrence of the failure mode.

[0036] Through this step, the equipment status, environmental conditions and historical fault data are weighted and integrated to obtain the corrected fault probability Sigmoid function. It is introduced to limit false positives, ensuring that the system only marks devices as abnormal when there is a high probability, reducing unnecessary intervention.

[0037] Implementation method 1: In this implementation method, the car washing machine is equipped with a series of sensors and monitoring equipment to comprehensively collect the operating data of each key component of the equipment. These data include multiple physical parameters such as current, voltage, temperature, pressure, and operating time, which are obtained in real time through sensors installed at various important positions of the car washing machine. These sensors continuously collect equipment operating status data at a high frequency (for example, every second or shorter time intervals) to ensure that the system can quickly respond to changes in equipment status.

[0038] The data is transmitted to the central control system or data processing module via wireless or wired communication. After preliminary processing, the data is transmitted to the monitoring platform in real time, and the system can monitor the working status of the equipment at any time. Each piece of collected data is automatically timestamped to ensure that the data can maintain accurate order in the time series. This provides a critical time reference for subsequent data analysis and decision-making, helping to identify potential risks as soon as equipment failure occurs.

[0039] As the car washing machine operates for a long time, the working state and operating environment of the equipment will change. For example, when the equipment is working in a high temperature environment, the normal range of its temperature change may be different from that in a low temperature environment. In order to avoid false positives and false negatives caused by traditional static thresholds, this implementation introduces an adaptive threshold algorithm to dynamically adjust the threshold range of each key parameter based on historical data and real-time operation data.

[0040] First, the system performs a preliminary analysis based on historical data and sets initial thresholds for each key parameter of the device. For example, motor temperature typically fluctuates within a certain range, and exceeding a certain threshold may indicate overheating, while current fluctuations reflect changes in motor load. Then, as the device continues to operate, the system dynamically adjusts based on real-time data and the device's operating environment. For example, when the device is operating under high load or high temperature, the system automatically adjusts the temperature and current thresholds to suit the current operating conditions. The dynamic adjustment process is calculated in real time by the algorithm and automatically updated according to the set strategy.

[0041] In addition, the adaptive threshold algorithm can also adjust the threshold according to different working stages. For example, when the device is started, some parameters may have short-term fluctuations, which is usually not a sign of failure. The system can set a loose threshold range to avoid false alarms. When the device enters a stable working state, the threshold can be more stringent to accurately capture minor anomalies in the device state changes.

[0042] After data collection and real-time processing, the system will compare each parameter value with the adjusted threshold. If a parameter exceeds the set normal working range, the system will immediately issue an alarm. For example, if the current of the motor suddenly increases, or the temperature rises rapidly in a short period of time, the system will notify the maintenance personnel through the control platform, SMS or email to indicate possible equipment failure or parts that need repair.

[0043] To reduce false alarms, the system will combine multiple key parameters to judge abnormalities. For example, if the increase in current and the sharp change in temperature occur at the same time, the system may judge it as a high-risk state of motor overload or failure, and a high-priority alarm will be triggered immediately. The system's early warning does not only rely on the single limit exceeding of a certain parameter, but also comprehensively considers the changing trends of multiple parameters, which enhances the accuracy of fault judgment.

[0044] When the system detects an abnormal state, it will classify the abnormality type according to the preset level. Minor abnormalities may only require regular inspections, while serious abnormalities require immediate equipment shutdown and repair. The system will record the abnormality in the background database for maintenance personnel to view and analyze. The system will also provide a detailed abnormality log, recording the cause of each equipment failure, the time of occurrence, the duration, and the response measures taken. In this way, subsequent maintenance personnel can evaluate the equipment status based on this data and quickly troubleshoot similar problems in the future.

[0045] Through these methods, the system not only improves the fault warning capability of the equipment, but also makes the maintenance of the equipment more efficient and intelligent. Adaptive threshold adjustment technology can minimize false alarms and missed alarms, reduce unnecessary inspections, and avoid excessive operating costs caused by frequent inspections.

[0046] Implementation method 2: In this implementation method, the system evaluates the status of the equipment by comprehensively considering multiple operating parameters of the equipment through multi-dimensional data fusion. Unlike traditional single characteristic parameter analysis, this implementation method combines multiple key parameters of the equipment (such as current, temperature, pressure, etc.) for collaborative analysis. By processing these data, the system can more accurately reflect the overall health of the equipment. For example, the temperature and current of the motor are closely related parameters. When the current changes, the temperature usually changes accordingly. If the current changes greatly but the temperature does not change, it may just be a normal fluctuation in the operation of the equipment; conversely, if the current change and the temperature increase occur at the same time, it is more likely to be a sign of failure.

[0047] Data fusion technology can integrate the data provided by multiple sensors into a comprehensive evaluation index, so that the system can make a more accurate judgment on the status of the equipment. For example, when the temperature, pressure and current all show a certain abnormal trend, the system will consider this to be a more serious fault signal, rather than just a false alarm of a single abnormal current or temperature.

[0048] Traditional fault detection systems often rely on the threshold judgment of a single parameter. This method is easily affected by environmental changes or fluctuations in the equipment itself, resulting in false alarms and missed alarms. However, this implementation significantly improves the accuracy of fault judgment by introducing a false alarm correction algorithm based on multi-dimensional joint judgment. The algorithm considers multiple characteristic parameters together. For example, when the motor current rises slightly, the system will not immediately trigger an alarm, but will comprehensively consider temperature, pressure and other data. If all these parameters show abnormalities, the system will judge it as a real fault.

[0049] In addition, the system uses algorithms to set different judgment rules to handle different situations. For example, for minor equipment fluctuations, the system will set a higher tolerance and will not trigger an alarm. For obvious equipment failure risks, the system will set a stricter threshold range to ensure that the failure can be caught in time. This multi-dimensional joint judgment can effectively reduce false alarms generated during equipment operation and avoid unnecessary interference.

[0050] After discovering equipment anomalies and correcting false alarms, the system will classify and handle them according to different anomaly types. Minor anomalies may recommend regular inspections by maintenance personnel, while more serious anomalies will trigger higher priority alarms and recommend immediate equipment shutdown inspections. All abnormal status and fault records will be stored in the background database for subsequent maintenance personnel to review and analyze. After each fault occurs, the system will automatically generate a fault report containing detailed information about the fault, the time of occurrence, abnormal characteristics, and the treatment measures taken.

[0051] Through the multi-dimensional joint judgment mechanism, the system can provide more accurate fault identification and significantly reduce the probability of false alarms and missed alarms, thereby reducing manual intervention and improving work efficiency. This method can effectively optimize the maintenance strategy of the car washing machine and extend the service life of the equipment.

[0052] Implementation method 3: The key to this implementation method is to train the historical operation data of the equipment through machine learning algorithms to build a fault prediction model. This model can analyze the operation trend of the equipment and identify the potential failure risk of the equipment. During the training stage of machine learning, the system will collect and analyze the historical data of the equipment, including the changes in multiple parameters such as current, voltage, temperature, and related data when the equipment fails. Through this data, the system can identify which change patterns are precursors to failures, and predict possible future failures based on these patterns.

[0053] Machine learning models can gradually improve the accuracy of fault prediction by continuously "learning" and optimizing. As the amount of data increases, the model will become more and more intelligent, able to identify more fault modes, predict different types of faults, and give corresponding warnings. For example, when the temperature continues to rise and is accompanied by current fluctuations, the machine learning model can recognize this pattern and judge it as a high-risk state of equipment overload or failure.

[0054] Once the model training is completed, the system will begin to predict and analyze real-time data. When the equipment is running, real-time data will be continuously input into the fault prediction model, and the system will evaluate the current risk level of the equipment based on the failure probability output by the model. If the system detects a potential fault signal, it will issue an early warning to the operator. The fault warning not only includes the estimated time of the fault, but also gives the possible fault type, such as motor failure, temperature control system failure, etc., and automatically adjusts the alarm level according to the risk level.

[0055] By combining machine learning models, the system can accurately predict when equipment failures will occur and remind maintenance personnel to take preventive measures to reduce the impact of equipment downtime due to failures. For example, the system can notify maintenance personnel in advance to check the motor to avoid downtime due to motor overheating.

[0056] As the equipment operation data continues to accumulate, the machine learning model will be continuously optimized and adjusted to improve the accuracy of fault prediction. The system can not only predict when the equipment will fail, but also provide intelligent decision support for maintenance personnel. Based on the results of the model analysis, the system will provide maintenance suggestions and automatically select the best maintenance plan based on the specific situation of the equipment. Through intelligent decision-making, the system can help maintenance personnel take measures at the most appropriate time, reduce equipment downtime, and improve the service efficiency of the car wash machine.

[0057] The combination of machine learning and self-optimization technology makes the maintenance of car wash machines more efficient and intelligent, which can significantly improve the operating reliability of the equipment, reduce maintenance costs, and extend the service life of the equipment.

[0058] Through adaptive threshold adjustment, multi-dimensional data fusion and machine learning technology, the intelligent early warning system of the present invention can greatly improve the accuracy of fault warnings and avoid the problems of false alarms and missed alarms caused by traditional static threshold judgments. The system dynamically adjusts the threshold according to the real-time working status and environmental conditions of the equipment, and comprehensively considers multiple key parameters to ensure that the early warning signal is only triggered when a real fault occurs, reducing unnecessary alarms and more accurately identifying potential fault risks. In addition, the continuous learning and optimization of historical data by the machine learning model enables the system to accurately predict the timing and type of equipment failures, further improving the reliability and accuracy of the early warning.

[0059] Through accurate fault warning and intelligent maintenance, the system of the present invention can detect potential equipment failure risks in advance, thereby avoiding downtime and high maintenance costs caused by major equipment failures. Through early warning, fault prediction and real-time monitoring, the system helps maintenance personnel take preventive measures in a timely manner, effectively reduces the occurrence of sudden equipment failures, and ensures the continuous and stable operation of the equipment. This not only significantly reduces equipment downtime and improves work efficiency, but also avoids equipment being in a high load or fault state for a long time, slows down the aging of the equipment, thereby extending the service life of the equipment and improving the overall economic benefits of the enterprise.

[0060] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0061] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0062] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0063] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean 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 application.

[0064] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0065] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0066] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0067] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0068] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0069] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent early warning car washing machine, characterized in that: It includes data acquisition module, data processing and feature extraction module, anomaly detection module, fault prediction and warning module, and false alarm correction and verification module: The data collection module collects the operation data of the car washing machine through preset sensors and monitoring equipment; The data processing and feature extraction module processes the collected data in real time, removes noise through filtering algorithms, and extracts key feature parameters related to the equipment operation status; The anomaly detection module uses an adaptive threshold judgment algorithm to determine whether each key feature parameter exceeds the normal working range based on historical data and set thresholds; The fault prediction and early warning module uses a fault prediction algorithm and a machine learning model to train historical fault data, predict potential fault types and occurrence probabilities, and generate corresponding early warning signals; The false alarm correction and verification module uses a false alarm correction algorithm based on multi-dimensional joint judgment to further verify the warning results, and performs weighted analysis on the abnormal detection results according to the real-time equipment status, environmental conditions and historical fault patterns to reduce the probability of false alarms and missed alarms. The specific steps are as follows: The real-time collected equipment data is scored in multiple dimensions to evaluate the current operating status of the equipment. The specific evaluation formula is as follows: , where is the comprehensive status score of the device. is the number of features involved in scoring, It is The weight of each feature in the comprehensive score, It is The real-time collection value of characteristic parameters, It is Normal working threshold of each feature; Environmental factors are introduced to make weighted adjustments to the equipment status score. The calculation formula for the adjustment factor of environmental conditions is as follows: , where is the weighted adjustment factor for environmental conditions, is the current environmental condition, is the standard normal environmental condition for the equipment, is the adjustment factor, is the nonlinear adjustment index.

2. The intelligent early warning car washing machine according to claim 1 is characterized in that: The specific steps for collecting the operation data of the car wash machine through the preset sensors and monitoring equipment are as follows: Collect key operating data of equipment in real time through sensors; To ensure data accuracy, a timestamp is added to each piece of data so that the order of the data can be maintained during subsequent analysis; Screen the collected data, remove errors and outliers, and ensure data quality; The collected data is stored in a data center or cloud server to ensure the real-time and integrity of data transmission.

3. The intelligent early warning car washing machine according to claim 1 is characterized in that: The specific steps for real-time processing of the collected data, removing noise through filtering algorithms, and extracting key characteristic parameters related to the equipment operating status are as follows: Use filtering algorithms to remove noise caused by external interference to ensure the smoothness and accuracy of data; By eliminating outliers and smoothing data fluctuations, we ensure that data is not affected by instantaneous interference; Extract key characteristic parameters reflecting the health status of the equipment from the processed data; The extracted feature data are standardized so that different feature parameters have the same dimension, which is convenient for subsequent analysis.

4. The intelligent early warning car washing machine according to claim 1 is characterized in that: Based on historical data and set thresholds, the specific steps of using the adaptive threshold judgment algorithm to determine whether each key characteristic parameter exceeds the normal working range are as follows: By analyzing historical operation data, a normal working status model of the equipment is established to provide a basis for setting thresholds; Set thresholds based on historical data and adjust them in real time through adaptive algorithms to adapt to different operating conditions; Compare real-time data with set thresholds. When the threshold is exceeded, it is marked as an abnormal state and an alarm is triggered. Abnormal conditions are marked and classified in a timely manner so that different levels of processing measures and feedback can be taken to ensure the normal operation of the equipment.

5. The intelligent early warning car washing machine according to claim 1 is characterized in that: The specific steps of using the fault prediction algorithm and the machine learning model to train historical fault data, predict potential fault types and occurrence probabilities, and generate corresponding warning signals are as follows: Before fault prediction, the historical fault data is preprocessed and feature engineered to provide input data for the machine learning model. By analyzing the operating status of the equipment before and after the fault, the factors that potentially affect the occurrence of the fault are extracted. The formula is as follows: , where is the original fault data matrix, is the processed data matrix, It is a data standardization operation. is a feature extraction operation; Using processed data The machine learning model is trained with labeled data. The model learns the characteristic performance of the device in normal and faulty states to predict the potential fault types and their probability of occurrence in the future. The formula is as follows , where is the optimal model parameter set obtained by minimizing the mean square error loss function, The model predicts The failure probability of a sample, It is the first in the historical fault data The true fault labels of samples, is the number of samples, is to find the parameters that minimize the loss function ; The model obtained through training , predict the newly collected real-time data to obtain the probability of equipment failure. The probability value will provide a basis for the subsequent early warning signal generation. The probability value calculation formula is: , where is the predicted probability of failure occurrence, is the newly collected real-time device data matrix, is the bias term, is the transpose of the matrix, yes Activation function; According to the probability of failure , according to the set threshold, it is determined whether to generate an early warning signal. The early warning signal provides a basis for subsequent decision-making. The judgment formula is as follows: , where It is a warning signal. If the predicted probability is greater than or equal to the threshold, it is 1 to trigger a warning, otherwise it is 0 and no warning is triggered. is the set fault occurrence probability threshold.

6. The intelligent early warning car washing machine according to claim 1, characterized in that: The abnormal detection results are corrected in combination with historical fault data. The correction factor calculation formula of historical fault mode is as follows: , where is the correction factor for the historical failure mode, It is The weight coefficient of the failure mode, It is The historical probability of occurrence of a failure mode, It is Equipment status score under various failure modes, is the total number of failure modes; Finally, combined with the comprehensive status score of the equipment , weighted adjustment factors for environmental conditions and correction factors for historical failure modes , comprehensively correct the warning results and obtain the final abnormal judgment probability. The calculation formula is as follows: , where is the corrected abnormality judgment probability, is the adjustment factor, is the natural base, is the historical probability of occurrence of the failure mode.

Images