Collection sound box fault detection method and terminal
By collecting key attribute information in the receipt speaker and using the comprehensive fault detection model for automatic detection, the problem of untimely fault handling in the existing technology is solved, the accuracy and efficiency of fault detection are improved, the operation cost is reduced, and the user experience is improved. .
Patent Information
- Application Number
- CN202411018346.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-08
AI Technical Summary
The lack of effective fault detection and early warning solutions in the existing technology, resulting in untimely handling of receipt speakers, increasing operating costs and affecting user experience, and weakening the market's trust in product quality and reliability.
A payment speaker failure detection method and terminal are adopted. By collecting key attribute information of historical preset time periods, inputting a comprehensive fault detection model for fault detection. The detection dimensions include battery life, payment delay time, terminal disconnection times, network The signal strength and memory usage are early warning based on the detection results, and automatic fault detection is performed using the Hidden Markov model, time series analysis model, statistical model, regression model and classification model.
It realizes automatic fault detection without user-active reporting, improves fault detection accuracy and comprehensiveness, shortens fault processing time, reduces manual intervention, reduces operating costs, and improves user trust and satisfaction with products.
Smart Images

Figure CN118921616A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault detection, and in particular to a method and a terminal for detecting faults of a cash register speaker. Background Art
[0002] The payment speaker is a voice announcement device specially designed for commercial transactions. It can receive and announce payment information obtained by scanning the payment code.
[0003] In previous practices, existing technologies mainly rely on users to actively report faulty terminals to locate problems. This method can be used when the number of terminals is small, but as the number of payment speakers shipped out of the warehouse continues to increase, this traditional fault handling method has gradually become inadequate. Not only does it increase operating costs, but the failure to handle the failure in a timely manner creates a vacuum period for users to use the product, and also has a negative impact on user experience. Frequent reports of product failures by users and the inconvenience in handling the process have gradually weakened the market's trust in product quality and reliability. If this situation continues, it will seriously affect the brand image and market competitiveness.
[0004] In summary, there is currently a lack of more effective fault detection and early warning solutions for payment speakers. Summary of the invention
[0005] The technical problem to be solved by the present invention is to provide a method and a terminal for detecting faults of a cash register speaker, which can realize more effective fault detection and early warning.
[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0007] A method for detecting a fault of a cash collection speaker comprises the following steps:
[0008] Collect key attribute information of payment collection speakers in a preset historical time period according to a preset frequency;
[0009] Input the key attribute information into a comprehensive fault detection model for fault detection, and output a detection result, wherein the detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength, and memory usage;
[0010] An early warning is issued based on the detection result.
[0011] In order to solve the above technical problems, another technical solution adopted by the present invention is:
[0012] A cash register speaker fault detection terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:
[0013] Collect key attribute information of payment collection speakers in a preset historical time period according to a preset frequency;
[0014] Input the key attribute information into a comprehensive fault detection model for fault detection, and output a detection result, wherein the detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength, and memory usage;
[0015] An early warning is issued based on the detection result.
[0016] The beneficial effects of the present invention are as follows: key attribute information of the payment collection speaker in a preset historical time period collected at a preset frequency is input into a comprehensive fault detection model for fault detection, and a detection result is output. The detection dimensions of the comprehensive fault detection model include battery life, payment collection delay time, number of terminal disconnections, network signal strength and memory usage. Finally, an early warning is issued based on the detection result. There is no need for users to actively report fault conditions. The comprehensive fault detection model can be used for automatic fault detection. The comprehensive fault detection model performs fault detection from multiple detection dimensions, which improves the fault detection accuracy and comprehensiveness of the payment collection speaker, and also improves the detection efficiency, thereby achieving more effective fault detection and early warning. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 A flowchart of a method for detecting a fault of a cash collection speaker according to an embodiment of the present invention;
[0018] Figure 2 The present invention is a schematic structural diagram of a payment speaker fault detection terminal according to an embodiment of the present invention. DETAILED DESCRIPTION
[0019] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.
[0020] Please refer to Figure 1 , a method for detecting a fault of a cash collection speaker, comprising the steps of:
[0021] Collect key attribute information of payment collection speakers in a preset historical time period according to a preset frequency;
[0022] Input the key attribute information into a comprehensive fault detection model for fault detection, and output a detection result, wherein the detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength, and memory usage;
[0023] An early warning is issued based on the detection result.
[0024] From the above description, it can be seen that the beneficial effects of the present invention are: the key attribute information of the payment speaker in the historical preset time period collected according to the preset frequency is input into the comprehensive fault detection model for fault detection, and the detection results are output. The detection dimensions of the comprehensive fault detection model include battery life, payment delay time, terminal disconnection times, network signal strength and memory usage. Finally, an early warning is issued based on the detection results. There is no need for users to actively report the fault. The comprehensive fault detection model can be used for automatic fault detection. The comprehensive fault detection model performs fault detection from multiple detection dimensions, which improves the fault detection accuracy and comprehensiveness of the payment speaker, and also improves the detection efficiency, thereby achieving more effective fault detection and early warning.
[0025] Furthermore, after collecting the key attribute information of the payment collection speaker in the historical preset time period according to the preset frequency, it also includes:
[0026] The key attribute information is subjected to data cleaning, denoising and conversion operations to obtain preprocessed key attribute information.
[0027] From the above description, it can be seen that data cleaning, denoising and conversion operations on key attribute information can eliminate invalid and abnormal data, improve data quality, and ensure the accuracy of subsequent fault detection.
[0028] Further, the key attribute information includes battery power and battery voltage;
[0029] The comprehensive fault detection model includes a hidden Markov model;
[0030] The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes:
[0031] The hidden Markov model is used to predict the life of the battery power and the battery voltage to obtain a predicted battery life, and the predicted battery life is compared with a first preset value to obtain a battery life detection result.
[0032] From the above description, we can know that the hidden Markov model is used to describe a Markov process with implicit unknown parameters. It has good flexibility, interpretability and robustness. The hidden Markov model is used to predict the battery power and battery voltage for life detection, which can timely detect abnormal battery life of the payment speaker.
[0033] Furthermore, the key attribute information includes the collection time and the broadcast time;
[0034] The comprehensive fault detection model includes a time series analysis model;
[0035] The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes:
[0036] Input the collection time and the broadcast time into a time series analysis model for prediction to obtain a predicted collection delay time, and determine the actual collection delay time based on the most recent collection time and broadcast time among the collection time and broadcast time;
[0037] The predicted payment collection delay time is compared with the actual payment collection delay time to obtain a payment collection delay time detection result.
[0038] From the above description, it can be seen that the time series analysis model uses statistical methods to predict the future based on past changing trends, which is usually in line with the law of development of things and focuses on the impact of periodic changes on specific time points. It is more accurate and uses the time series analysis model to perform fault detection on the collection delay time, thereby achieving more accurate fault detection of the collection delay time of the collection speaker.
[0039] Furthermore, the key attribute information includes a network disconnection event;
[0040] The comprehensive fault detection model includes a statistical model;
[0041] The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes:
[0042] The network disconnection event is input into the statistical model to obtain the number of disconnections, and the number of disconnections is compared with a second preset value to obtain a terminal disconnection number detection result.
[0043] From the above description, it can be seen that too many disconnections will cause the cash collection speaker to fail to work properly. The network disconnection events are counted through a statistical model to obtain the number of disconnections, and the number of disconnections is compared with the second preset value to ensure that the faulty cash collection speaker is discovered in time.
[0044] Further, the key attribute information includes signal strength and device location data;
[0045] The comprehensive fault detection model includes a regression model and a classification model;
[0046] The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes:
[0047] The signal strength and the device location data are input into the regression model for prediction to obtain the predicted signal strength, and the predicted signal strength is input into the classification model for abnormality judgment to obtain a network signal strength detection result.
[0048] From the above description, we can see that the regression model can accurately predict the signal strength, and the classification model can accurately determine whether the signal strength is abnormal. Therefore, the regression model and the classification model are used to perform fault detection on the signal strength to ensure the reliability of network signal strength detection.
[0049] Furthermore, the key attribute information includes memory occupancy rate;
[0050] The comprehensive fault detection model includes a memory occupancy detection model;
[0051] The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes:
[0052] The memory occupancy rate is input into the memory occupancy detection model and compared with a third preset value to obtain a memory occupancy detection result.
[0053] From the above description, it can be seen that since the memory occupancy rate of the cash collection speaker continues to rise and exceeds the preset value, it will affect the normal use of the cash collection speaker. The memory occupancy rate is input into the memory occupancy detection model and compared with the third preset value to obtain the memory occupancy detection result, so as to monitor the memory occupancy and realize effective cash collection speaker fault detection.
[0054] Furthermore, the early warning based on the detection result includes:
[0055] If the detection result of any detection dimension in the detection results is abnormal, a fault warning is issued to the payment speaker based on the abnormal detection result.
[0056] From the above description, it can be seen that as long as the detection result of any detection dimension is abnormal, it is determined that the payment speaker is faulty and a fault warning is required so as to promptly notify the staff to handle the faulty payment speaker.
[0057] Furthermore, the fault warning of the payment receiving speaker based on the abnormal detection result includes:
[0058] Classify the faults according to the abnormal detection results to obtain the classified faults;
[0059] An early warning method is determined according to the classified faults, and the early warning method is used to issue a fault early warning to the payment collection speaker.
[0060] From the above description, it can be seen that by classifying faults, different faults after different classifications are warned in different ways, so that the faults can be handled in a targeted manner, ensuring that faults with high impact can be handled first, and achieving more effective fault detection and warning.
[0061] Please refer to Figure 2Another embodiment of the present invention provides a cash register speaker fault detection terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements each step of the above-mentioned cash register speaker fault detection method when executing the computer program.
[0062] The above-mentioned cash collection speaker fault detection method and terminal of the present invention can be applied to the cash collection speaker, and the following is described through specific implementation methods:
[0063] Please refer to Figure 1 , Embodiment 1 of the present invention is:
[0064] A method for detecting a fault of a cash collection speaker comprises the following steps:
[0065] S1. Collect key attribute information of the payment collection speaker in a preset historical time period according to a preset frequency.
[0066] Among them, the key attribute information includes battery power, battery voltage, collection time, broadcast time, network disconnection events, signal strength, device location data and memory usage.
[0067] In an optional implementation, the preset frequency is once every half an hour, and the historical preset time period is from the current collection time to the power-on time of the payment speaker, to ensure the timeliness of the data. In an optional implementation, the key attribute information uploaded by the payment speaker is received through MQTT (Message Queuing Telemetry Transport) to ensure the real-time and reliability of data transmission, and the received key attribute information is forwarded to the Kafka cluster for storage, using the high throughput and distributed characteristics of Kafka to provide solid support for subsequent data processing.
[0068] S2. Perform data cleaning, denoising and conversion operations on the key attribute information to obtain pre-processed key attribute information, thereby eliminating invalid and abnormal data and improving data quality.
[0069] In an optional implementation, the preprocessed key attribute information may be securely stored in a distributed database.
[0070] S3, inputting the key attribute information into a comprehensive fault detection model for fault detection, and outputting the detection result. The detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength and memory usage, specifically including S31-S36:
[0071] The comprehensive fault detection model includes a hidden Markov model, a time series analysis model, a statistical model, a regression model, a classification model and a memory occupancy detection model. In an optional implementation, the detection dimension of the comprehensive fault detection model is determined in advance according to the historical fault type of the cashier speaker.
[0072] S31. Use the hidden Markov model to predict the battery power and battery voltage to obtain a predicted battery life, and compare the predicted battery life with a first preset value to obtain a battery life detection result.
[0073] The hidden Markov model can predict the time-varying sequence of battery status according to the input battery power and battery voltage, and estimate the expected life of the battery according to the sequence and the battery usage. The battery status includes new, healthy, degraded and faulty.
[0074] The predicted battery life is compared with a first preset value to obtain a battery life detection result, specifically:
[0075] If the predicted battery life is lower than the first preset value, it is determined that the battery life detection result is abnormal.
[0076] S32: Input the collection time and the broadcast time into the time series analysis model for prediction to obtain the predicted collection delay time, and determine the actual collection delay time according to the most recent collection time and the broadcast time.
[0077] Specifically, the historical collection time and broadcast time except the most recent collection time and broadcast time are input into the time series analysis model for prediction to obtain the predicted collection delay time, and the actual collection delay time is determined based on the most recent collection time and broadcast time.
[0078] S33. Compare the predicted payment collection delay time with the actual payment collection delay time to obtain a payment collection delay time detection result.
[0079] Specifically, if the actual payment collection delay time exceeds the predicted payment collection delay time, the payment collection delay time detection result is determined to be abnormal.
[0080] S34: Input the network disconnection event into the statistical model to obtain the number of disconnections, and compare the number of disconnections with a second preset value to obtain a terminal disconnection number detection result.
[0081] The comparing the number of disconnection times with a second preset value to obtain a terminal disconnection times detection result is specifically:
[0082] If the number of disconnections is greater than a second preset value, the terminal disconnection detection result is determined to be abnormal. The second preset value is a reasonable number of terminal disconnections.
[0083] S35. Input the signal strength and the device location data into the regression model for prediction to obtain the predicted signal strength, and input the predicted signal strength into the classification model for abnormality judgment to obtain a network signal strength detection result.
[0084] The classification model determines whether the predicted signal strength is normal or abnormal, where abnormality includes severe probability abnormality, high probability abnormality, medium probability abnormality and medium probability abnormality, so as to obtain a network signal strength detection result.
[0085] S36: Input the memory occupancy rate into the memory occupancy detection model and compare it with a third preset value to obtain a memory occupancy detection result.
[0086] Specifically, the memory occupancy rate is input into the memory occupancy detection model, and if the memory occupancy rate is greater than a third preset value, it is determined that the memory occupancy detection result is abnormal.
[0087] In an optional implementation, it also includes:
[0088] The performance of each model in the comprehensive fault detection model is regularly evaluated to obtain evaluation results, and each model is optimized according to the evaluation results, so as to continuously optimize the model algorithm and parameter settings and improve the detection accuracy and efficiency.
[0089] S4. Issue an early warning based on the detection result.
[0090] Specifically, if the detection result of any detection dimension in the detection result is abnormal, a fault warning is performed on the payment receiving speaker based on the abnormal detection result, specifically including S41-S42:
[0091] S41. Classify the faults according to the abnormal detection results to obtain classified faults.
[0092] Specifically, for abnormal battery life detection results, if the predicted battery life is lower than a first preset percentage of a first preset value, it is determined as a level one fault; if the predicted battery life is lower than a second preset percentage of the first preset value, it is determined as a level two fault; if the predicted battery life is lower than a third preset percentage of the first preset value, it is determined as a level three fault; if the predicted battery life is lower than a fourth preset percentage of the first preset value, it is determined as a level four fault; in an optional embodiment, the first preset percentage, the second preset percentage, the third preset percentage and the fourth preset percentage are 20%, 40%, 60% and 80% respectively.
[0093] For abnormal collection delay time detection results, if the actual collection delay time exceeds the first preset multiple of the predicted collection delay time and exceeds the first preset time, it is determined to be a level one fault; if the actual collection delay time exceeds the second preset multiple of the predicted collection delay time, but does not exceed the first preset multiple, it is determined to be a level two fault; if the actual collection delay time exceeds the third preset multiple of the predicted collection delay time, but does not exceed the second preset multiple, it is determined to be a level three fault; if the actual collection delay time exceeds the fourth preset multiple of the predicted collection delay time, but does not exceed the third preset multiple, it is determined to be a level four fault; in an optional implementation, the first preset multiple, the second preset multiple, the third preset multiple and the fourth preset multiple are 5 times, 3 times, 2 times and 1.5 times respectively, and the first preset time is 5 minutes.
[0094] For the abnormal terminal offline detection result, if the offline number is greater than the fifth preset multiple of the second preset value, it is determined to be a level one fault; if the offline number is greater than the sixth preset multiple of the second preset value and does not exceed the fifth preset multiple, it is determined to be a level two fault; if the offline number is greater than the seventh preset multiple of the second preset value and does not exceed the sixth preset multiple, it is determined to be a level three fault; if the offline number is greater than the eighth preset multiple of the second preset value and does not exceed the seventh preset multiple, it is determined to be a level four fault; in one embodiment, the fifth preset multiple is 3 times, the sixth preset multiple is 2 times, the seventh preset multiple is 1.5 times, and the eighth preset multiple is 1 times.
[0095] For abnormal network signal strength detection results, if the classification model judges it as a serious probability abnormality, it is determined to be a level one fault; if the classification model judges it as a high probability abnormality, it is determined to be a level two fault; if the classification model judges it as a medium probability abnormality, it is determined to be a level three fault; if the classification model judges it as a low probability abnormality, it is determined to be a level four fault.
[0096] For the abnormal memory occupancy detection result, if the memory occupancy rate is greater than the ninth preset multiple of the third preset value and the duration exceeds the second preset time, it is determined as a level one fault; if the memory occupancy rate is greater than the tenth preset multiple of the third preset value but does not reach the ninth preset multiple, it is determined as a level two fault; if the memory occupancy rate is greater than the eleventh preset multiple of the third preset value but does not reach the tenth preset multiple, it is determined as a level three fault; if the memory occupancy rate is greater than the twelfth preset multiple of the third preset value but does not reach the eleventh preset multiple, it is determined as a level four fault; in one embodiment, the ninth preset multiple is 1.2 times, the tenth preset multiple is 1.15 times, the eleventh preset multiple is 1.1 times, the twelfth preset multiple is 1 times, and the second preset time is 1 hour.
[0097] S42: determining an early warning method according to the classified faults, and using the early warning method to issue a fault early warning to the payment collection speaker.
[0098] Specifically, if the fault after classification is a level one fault, the early warning method is determined to be notification by telephone, text message and email, and the emergency response process is initiated. If the fault after classification is a level two fault, the early warning method is determined to be notification by email and instant messaging, and preliminary fault troubleshooting is performed. If the fault after classification is a level three fault, the early warning method is determined to be notification by email, and the fault-related information is recorded for regular fault troubleshooting. If the fault after classification is a level four fault, the early warning method is determined to be notification through internal communication tools, and the fault-related information is recorded for regular fault troubleshooting.
[0099] The above-mentioned cash register speaker fault detection method of the present invention combines data analysis technology and machine learning models to accurately identify potential faulty devices, ensuring that the problem is resolved before the user notices it, and ensuring that once an abnormality is detected, the response mechanism can be triggered immediately, greatly shortening the fault handling time. In addition, by reducing the impact of cash register speaker failures on user operations, the user's trust and satisfaction with the product is improved, which not only reduces the need for manual intervention, but also optimizes the maintenance process and reduces operating costs.
[0100] Please refer to Figure 2 , Embodiment 2 of the present invention is:
[0101] A cash collection speaker fault detection terminal comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the cash collection speaker fault detection method in the first embodiment is implemented.
[0102] In summary, the present invention provides a method and terminal for detecting faults of a cash collection speaker, which inputs key attribute information of a cash collection speaker in a preset historical time period collected at a preset frequency into a comprehensive fault detection model for fault detection and outputs a detection result. The detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength and memory usage. Finally, an early warning is issued based on the detection result. There is no need for the user to actively report the fault. The comprehensive fault detection model can be used for automatic fault detection, and the comprehensive fault detection model performs fault detection from multiple detection dimensions, thereby improving the fault detection accuracy and comprehensiveness of the cash collection speaker and also improving the detection efficiency, thereby achieving more effective fault detection and early warning. In addition, as long as the detection result of any detection dimension is abnormal, it is determined that the cash collection speaker has a fault and a fault early warning is required so as to promptly notify the staff to handle the faulty cash collection speaker. Moreover, by grading the faults, different graded faults are given different early warning methods for fault early warning, so that the faults can be handled in a targeted manner to ensure that faults with a high degree of impact can be handled first, thereby achieving more effective fault detection and early warning.
[0103] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A method for detecting a fault in a cash register speaker, characterized in that: Includes steps: Collect key attribute information of payment collection speakers in a preset historical time period according to a preset frequency; Input the key attribute information into a comprehensive fault detection model for fault detection, and output a detection result, wherein the detection dimensions of the comprehensive fault detection model include battery life, payment delay time, number of terminal disconnections, network signal strength, and memory usage; An early warning is issued based on the detection result.
2. A method for detecting a fault of a cash register speaker according to claim 1, characterized in that: After collecting the key attribute information of the payment collection speakers in the historical preset time period according to the preset frequency, it also includes: The key attribute information is subjected to data cleaning, denoising and conversion operations to obtain preprocessed key attribute information.
3. A method for detecting a fault of a cash register speaker according to claim 2, characterized in that: The key attribute information includes battery power and battery voltage; The comprehensive fault detection model includes a hidden Markov model; The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes: The hidden Markov model is used to predict the life of the battery power and the battery voltage to obtain a predicted battery life, and the predicted battery life is compared with a first preset value to obtain a battery life detection result.
4. A method for detecting a fault of a cash register speaker according to claim 2, characterized in that: The key attribute information includes the payment time and the broadcast time; The comprehensive fault detection model includes a time series analysis model; The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes: Input the collection time and the broadcast time into a time series analysis model for prediction to obtain a predicted collection delay time, and determine the actual collection delay time based on the most recent collection time and broadcast time among the collection time and broadcast time; The predicted payment collection delay time is compared with the actual payment collection delay time to obtain a payment collection delay time detection result.
5. A method for detecting a fault of a cash register speaker according to claim 2, characterized in that: The key attribute information includes network disconnection events; The comprehensive fault detection model includes a statistical model; The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes: The network disconnection event is input into the statistical model to obtain the number of disconnections, and the number of disconnections is compared with a second preset value to obtain a terminal disconnection number detection result.
6. A method for detecting a fault of a cash register speaker according to claim 2, characterized in that: The key attribute information includes signal strength and device location data; The comprehensive fault detection model includes a regression model and a classification model; The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes: The signal strength and the device location data are input into the regression model for prediction to obtain the predicted signal strength, and the predicted signal strength is input into the classification model for abnormality judgment to obtain a network signal strength detection result.
7. A method for detecting a fault of a cash register speaker according to claim 2, characterized in that: The key attribute information includes memory occupancy rate; The comprehensive fault detection model includes a memory occupancy detection model; The inputting the key attribute information into the comprehensive fault detection model to perform fault detection, and outputting the detection result includes: The memory occupancy rate is input into the memory occupancy detection model and compared with a third preset value to obtain a memory occupancy detection result.
8. A method for detecting a fault of a cash register speaker according to claim 1, characterized in that: The early warning based on the detection result includes: If the detection result of any detection dimension in the detection results is abnormal, a fault warning is issued to the payment speaker based on the abnormal detection result.
9. A method for detecting a fault of a cash register speaker according to claim 8, characterized in that: The fault warning of the payment receiving speaker based on the abnormal detection result includes: Classify the faults according to the abnormal detection results to obtain the classified faults; An early warning method is determined according to the classified faults, and the early warning method is used to issue a fault early warning to the payment collection speaker.
10. A terminal for detecting faults of a cash collection speaker, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, each step of the method for detecting a fault of a cash register speaker according to any one of claims 1 to 9 is implemented.
