Fault detection method and device in property service

Abstract

The invention discloses a fault detection method and device in property service, and the method comprises the steps: collecting sensing data in a target region through an Internet of Things sensing assembly, and uploading the sensing data to an analysis model; determining a health index corresponding to the sensing data by using the analysis model; determining a dynamic health threshold using the analysis model; determining a prediction analysis result according to the health index and the dynamic health threshold; the analysis model is used for analyzing sensing data, potential risks are explored, and faults are predicted. Automatic fault detection is realized, and the problem of dependence on manual inspection in the prior art is avoided. The risk can be predicted in advance; and whether a fault occurs or not is judged by combining a dynamic health threshold with an objective environment, so that the problem that a fixed threshold alarm mechanism is rigid is solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a fault detection method and device in property services. Background Art

[0002] Currently, intelligent property management services are based on Enterprise Resource Planning (ERP). ERP is an integrated management software system with work order management capabilities, enabling the management of inspections and maintenance work in the property management area.

[0003] Existing ERP systems trigger alarms based on fixed thresholds (e.g., when an elevator's operating time exceeds a set value), but they are unable to dynamically adjust thresholds or predict potential failures. In other words, existing ERP systems rely heavily on manual inspections and reactive maintenance, resulting in delayed fault response, high maintenance costs, and a high rate of sudden equipment failures.

[0004] Specifically, the current ERP system has the following problems.

[0005] Heavy reliance on manual inspections leads to low efficiency. Traditional property facility maintenance relies primarily on regular manual inspections with fixed inspection cycles (e.g., weekly), which fail to detect equipment anomalies in real time. Manual record-keeping is prone to errors, and paper work orders are difficult to trace. Inspection coverage is low during special periods such as nighttime and holidays. Sudden failures can be detected without early warning (e.g., a water pump bearing may be worn and broken before being discovered), resulting in high repair costs.

[0006] Fixed threshold alarm mechanisms are rigid. Existing systems typically set static thresholds to trigger alarms (e.g., an elevator will trigger an alarm if it runs continuously for 10 hours). However, different equipment ages and uses vary widely, making a unified threshold unscientific. Environmental factors (such as temperature and humidity) are not factored into threshold calculations, leading to a high false alarm rate (e.g., high temperatures causing sensors to falsely trigger). For example, a fire pump experienced reduced insulation performance due to increased humidity, but the system failed to dynamically adjust the current threshold, failing to warn of an eventual short circuit and burnout.

[0007] Data silos are a prominent problem, and predictive capabilities are lacking. Equipment operating data, energy consumption data, and maintenance records are scattered across different subsystems (e.g., elevator monitoring is independent of the power distribution system). This makes root cause analysis difficult (e.g., voltage fluctuations can cause an elevator outage, but the data from the two systems is unconnected). Historical data is not used for modeling, but only for retrospective analysis.

[0008] Resource scheduling relies on experience, resulting in delayed responses. Work order assignments are entirely manual, leading to mismatches between maintenance personnel skills and tasks (e.g., an electrician assigned to fix a mechanical problem); inappropriate routing (multiple trips to the same building); and subjective prioritization of urgent tasks. Summary of the Invention

[0009] The present invention provides a fault detection method and device in property services to achieve automated fault analysis and prediction.

[0010] In a first aspect, the present invention provides a fault detection method in property management services, comprising:

[0011] Using IoT sensor components to collect sensor data within the target area and upload the sensor data to the analysis model;

[0012] Determining a health index corresponding to the sensor data using the analysis model;

[0013] determining a dynamic health threshold using the analytical model;

[0014] A prediction analysis result is determined based on the health index and the dynamic health threshold.

[0015] Preferably, uploading the sensor data to the analysis model comprises:

[0016] The sensor data is preprocessed using an edge computing node, and the preprocessed sensor data is uploaded to the analysis model.

[0017] Preferably, it also includes:

[0018] Determine the original neural network;

[0019] Performing data training on the original neural network using historical data;

[0020] The trained original neural network is used as the analysis model.

[0021] Preferably, the determining the health index corresponding to the sensor data by using the analysis model includes:

[0022] extracting spatial and temporal features of the sensor data using the analysis model;

[0023] The health index is determined according to the spatial characteristics and the temporal characteristics.

[0024] Preferably, determining the dynamic health threshold using the analysis model includes:

[0025] Determine environmental data;

[0026] The dynamic health threshold is determined based on a Bayesian optimization algorithm and using the environmental data.

[0027] Preferably, determining the prediction analysis result according to the health index and the dynamic health threshold comprises:

[0028] When the health index does not reach the dynamic health threshold, determining a fault type and a fault probability based on the spatial characteristics and the temporal characteristics;

[0029] The fault type and the fault probability are used as the prediction analysis result.

[0030] Preferably, it also includes:

[0031] determining a maintenance work order based on the predictive analysis results;

[0032] Determine the work path of the maintenance resource and at least one backup personnel;

[0033] The allocation strategy is determined according to the maintenance resources and the work path.

[0034] In a second aspect, the present invention provides a fault detection device for property services, comprising:

[0035] A data acquisition module, configured to collect sensor data within a target area using IoT sensor components and upload the sensor data to an analysis model;

[0036] a data analysis module, configured to determine a health index corresponding to the sensor data using the analysis model;

[0037] a dynamic threshold module, configured to determine a dynamic health threshold using the analysis model;

[0038] The prediction analysis module is used to determine the prediction analysis result based on the health index and the dynamic health threshold.

[0039] In a third aspect, the present invention provides a readable medium comprising an execution instruction. When a processor of an electronic device executes the execution instruction, the electronic device executes any method described in the first aspect.

[0040] In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory storing execution instructions, wherein when the processor executes the execution instructions stored in the memory, the processor executes any method described in the first aspect.

[0041] This invention provides a fault detection method and device for property management services. This method uses analytical models to analyze sensor data, identify potential risks, and predict faults. This method achieves automated fault detection, avoiding the existing reliance on manual inspections. Risks can be predicted in advance. Furthermore, dynamic health thresholds combined with objective environmental factors determine whether a fault has occurred, addressing the rigidity of fixed-threshold alarm mechanisms.

[0042] The further effects of the above-mentioned non-conventional preferred embodiment will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the existing technical solutions, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A flowchart of a method for fault detection in property services provided by one embodiment of the present invention;

[0045] Figure 2 A flowchart of another method for fault detection in property services provided by an embodiment of the present invention;

[0046] Figure 3 A schematic structural diagram of a fault detection device for property management services provided by an embodiment of the present invention;

[0047] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] Currently, intelligent property management services are based on Enterprise Resource Planning (ERP). ERP is an integrated management software system with work order management capabilities, enabling the management of inspections and maintenance work in the property management area.

[0050] Existing ERP systems trigger alarms based on fixed thresholds (e.g., when an elevator's operating time exceeds a set value), but they are unable to dynamically adjust thresholds or predict potential failures. In other words, existing ERP systems rely heavily on manual inspections and reactive maintenance, resulting in delayed fault response, high maintenance costs, and a high rate of sudden equipment failures.

[0051] Specifically, the current ERP system has the following problems.

[0052] Heavy reliance on manual inspections leads to low efficiency. Traditional property facility maintenance relies primarily on regular manual inspections with fixed inspection cycles (e.g., weekly), which fail to detect equipment anomalies in real time. Manual record-keeping is prone to errors, and paper work orders are difficult to trace. Inspection coverage is low during special periods such as nighttime and holidays. Sudden failures can be detected without early warning (e.g., a water pump bearing may be worn and broken before being discovered), resulting in high repair costs.

[0053] Fixed threshold alarm mechanisms are rigid. Existing systems typically set static thresholds to trigger alarms (e.g., an elevator will trigger an alarm if it runs continuously for 10 hours). However, different equipment ages and uses vary widely, making a unified threshold unscientific. Environmental factors (such as temperature and humidity) are not factored into threshold calculations, leading to a high false alarm rate (e.g., high temperatures causing sensors to falsely trigger). For example, a fire pump experienced reduced insulation performance due to increased humidity, but the system failed to dynamically adjust the current threshold, failing to warn of an eventual short circuit and burnout.

[0054] Data silos are a prominent problem, and predictive capabilities are lacking. Equipment operating data, energy consumption data, and maintenance records are scattered across different subsystems (e.g., elevator monitoring is independent of the power distribution system). This makes root cause analysis difficult (e.g., voltage fluctuations can cause an elevator outage, but the data from the two systems is unconnected). Historical data is not used for modeling, but only for retrospective analysis.

[0055] Resource scheduling relies on experience, resulting in delayed responses. Work order assignments are entirely manual, leading to mismatches between maintenance personnel skills and tasks (e.g., an electrician assigned to fix a mechanical problem); inappropriate routing (multiple trips to the same building); and subjective prioritization of urgent tasks.

[0056] In view of this, the present invention provides a fault detection method in property services. Figure 1 FIG. 1 is a specific embodiment of the fault detection method in property management services provided by the present invention. In this embodiment, the method includes:

[0057] Step 101: Use the Internet of Things sensor components to collect sensor data in the target area and upload the sensor data to the analysis model.

[0058] It can be considered that the executor of the method in this embodiment is a fault detection system. This system combines hardware and software to operate in a collaborative manner to achieve a closed loop from data collection to data analysis. The Internet of Things sensor component is a part of the hardware equipment in the system. The Internet of Things sensor component mainly includes various sensors deployed in the target area and the Internet of Things communication module. The so-called target area is the object targeted by the fault detection and the area where the object is located. For example, to perform fault detection on a certain device, sensors can be deployed on the device and in the specific space where the device is located. Specifically, sensors can include vibration sensors (to monitor motor anomalies), infrared sensors (to detect overheating), smart meters (to monitor sudden changes in energy consumption), etc. After the sensors collect the sensor data, the sensor data is uploaded to the analysis model by the Internet of Things communication module. The Internet of Things communication module can support multi-protocol adaptation such as MQTT and Modbus to ensure access to devices from different manufacturers.

[0059] It should also be noted that before actually uploading the sensor data, the sensor data can be preprocessed using the edge computing node and uploaded to the analysis model. The preprocessing can specifically include filtering and normalization preprocessing.

[0060] Step 102: Determine the health index corresponding to the sensor data using the analysis model.

[0061] In this embodiment, the analysis model is a network model pre-established based on artificial intelligence technology. Specifically, an original neural network can be determined; the original neural network is trained using historical data; and the trained original neural network is used as the analysis model.

[0062] Structurally, the analysis model can adopt a hybrid model of a convolutional neural network (CNN) and a long short-term memory network (LSTM), that is, a CNN-LSTM network. The trained analysis model can learn from historical data, perform intelligent analysis on the sensor data, and discover whether there are certain potential anomalies in the sensor data. Then, these anomalies can be used to predict whether a failure will occur, what kind of failure will occur, and when the failure will occur. Specifically, the CNN network in the analysis model can be used to extract the spatial features of the sensor data, and the LSTM network can be used to extract the temporal features of the sensor data. The health index is determined based on the spatial features and the temporal features.

[0063] For example, the input layer of the analysis model can capture time series data through a sliding window (window size = 24 hours). The CNN layer can include three layers of convolution kernels (size = 5×1) to extract spatial features. The LSTM layer can capture long-term dependencies, that is, temporal features, through two layers of neurons (128 units). The output layer uses a Sigmoid activation function to output the probability of failure. For example, sensor data X1 = current fluctuation and X2 = temperature is input to the analysis model. Spatial features (such as abnormal current waveform patterns) are extracted through CNN, and temporal features (such as intensified vibration for three consecutive days) are captured by LSTM. The analysis model can output a quantitative health index.

[0064] This embodiment uses an analytical model to analyze sensor data, identify potential risks, and predict failures. This enables automated fault detection, avoiding the reliance on manual inspections found in existing technologies. Risks can be predicted in advance. Furthermore, the analytical model is trained using extensive historical data, enabling comprehensive and integrated analysis, resolving the data silo problem found in existing technologies.

[0065] Step 103: Determine a dynamic health threshold using the analysis model.

[0066] In other cases, existing detection methods also suffer from the rigidity of fixed threshold alarm mechanisms. Existing systems typically set static thresholds to trigger alarms (e.g., an elevator will trigger an alarm after 10 hours of continuous operation). However, different equipment ages and uses vary widely, making a unified threshold unscientific. Environmental factors (such as temperature and humidity) are not factored into threshold calculations, leading to a high false alarm rate (e.g., high temperatures causing sensors to trigger falsely).

[0067] Therefore, a dynamic health threshold is set in this embodiment, and the dynamic health threshold is used as the basis for judging whether the health index meets the standard. Specifically, environmental data can be determined; based on the Bayesian optimization algorithm, and using the environmental data, the dynamic health threshold is determined. In other words, the "health threshold" in this embodiment can be adjusted in a timely manner according to the objective environment to adapt to the different working conditions of the equipment in different environments, so as to more objectively judge whether there is a fault or not, and shield the impact of environmental changes on the sensor data. For example, if the air conditioning load is high in summer, the temperature alarm threshold can be adjusted from 70°C to 75°C. This will not affect the normal operation of the equipment, and at the same time take into account the analysis and judgment of the fault.

[0068] By setting dynamic health thresholds, the rigidity of the fixed threshold alarm mechanism is solved.

[0069] Step 104: Determine a prediction analysis result based on the health index and the dynamic health threshold.

[0070] By comparing the health index and the dynamic health threshold, the predictive analysis results of fault detection can be determined, that is, whether the equipment operating conditions represented by the sensor data are healthy, whether there are risks, and the possibility of failure under the current health index.

[0071] If the health index ranges from 0 to 100, with higher values indicating greater health, the dynamic health threshold is 60. If the actual health index falls below 60, it indicates a risk of failure, and personnel can be assigned for further action. This allows for troubleshooting before it actually occurs.

[0072] The above technical solution demonstrates the beneficial effects of this embodiment: It utilizes analytical models to analyze sensor data, identify potential risks, and predict failures. This enables automated fault detection, avoiding the reliance on manual inspections found in existing technologies. Risks can be anticipated in advance. Furthermore, dynamic health thresholds, combined with objective environmental factors, are used to determine whether a fault has occurred, addressing the rigidity of fixed threshold alarm mechanisms.

[0073] Figure 1 What is shown is only a basic embodiment of the method of the present invention. By performing certain optimization and expansion on this basis, other preferred embodiments of the method can be obtained.

[0074] like Figure 2 FIG. 1 is another specific embodiment of a fault detection method for property management services according to the present invention. This embodiment further describes the above embodiment. In this embodiment, the method includes the following steps:

[0075] Step 201: Use the Internet of Things sensor components to collect sensor data in the target area and upload the sensor data to the analysis model.

[0076] It can be considered that the executor of the method in this embodiment is a fault detection system. This system combines hardware and software to operate in a collaborative manner to achieve a closed loop from data collection to data analysis. The Internet of Things sensor component is a part of the hardware equipment in the system. The Internet of Things sensor component mainly includes various sensors deployed in the target area and the Internet of Things communication module. The so-called target area is the object targeted by the fault detection and the area where the object is located. For example, to perform fault detection on a certain device, sensors can be deployed on the device and in the specific space where the device is located. Specifically, sensors can include vibration sensors (to monitor motor anomalies), infrared sensors (to detect overheating), smart meters (to monitor sudden changes in energy consumption), etc. After the sensors collect the sensor data, the sensor data is uploaded to the analysis model by the Internet of Things communication module. The Internet of Things communication module can support multi-protocol adaptation such as MQTT and Modbus to ensure access to devices from different manufacturers.

[0077] It should also be noted that before actually uploading the sensor data, the sensor data can be preprocessed using the edge computing node and uploaded to the analysis model. The preprocessing can specifically include filtering and normalization preprocessing.

[0078] Step 202: Determine the health index corresponding to the sensor data using the analysis model.

[0079] In this embodiment, the analysis model is a network model pre-established based on artificial intelligence technology. Specifically, an original neural network can be determined; the original neural network is trained using historical data; and the trained original neural network is used as the analysis model.

[0080] Structurally, the analysis model can adopt a hybrid model of a convolutional neural network (CNN) and a long short-term memory network (LSTM), that is, a CNN-LSTM network. The trained analysis model can learn from historical data, perform intelligent analysis on the sensor data, and discover whether there are certain potential anomalies in the sensor data. Then, these anomalies can be used to predict whether a failure will occur, what kind of failure will occur, and when the failure will occur. Specifically, the CNN network in the analysis model can be used to extract the spatial features of the sensor data, and the LSTM network can be used to extract the temporal features of the sensor data. The health index is determined based on the spatial features and the temporal features.

[0081] For example, the input layer of the analysis model can capture time series data through a sliding window (window size = 24 hours). The CNN layer can include three layers of convolution kernels (size = 5×1) to extract spatial features. The LSTM layer can capture long-term dependencies, that is, temporal features, through two layers of neurons (128 units). The output layer uses a Sigmoid activation function to output the probability of failure. For example, sensor data X1 = current fluctuation and X2 = temperature is input to the analysis model. Spatial features (such as abnormal current waveform patterns) are extracted through CNN, and temporal features (such as intensified vibration for three consecutive days) are captured by LSTM. The analysis model can output a quantitative health index.

[0082] This embodiment uses an analytical model to analyze sensor data, identify potential risks, and predict failures. This enables automated fault detection, avoiding the reliance on manual inspections found in existing technologies. Risks can be predicted in advance. Furthermore, the analytical model is trained using extensive historical data, enabling comprehensive and integrated analysis, resolving the data silo problem found in existing technologies.

[0083] Step 203: Determine a dynamic health threshold using the analysis model.

[0084] In other cases, existing detection methods also suffer from the rigidity of fixed threshold alarm mechanisms. Existing systems typically set static thresholds to trigger alarms (e.g., an elevator will trigger an alarm after 10 hours of continuous operation). However, different equipment ages and uses vary widely, making a unified threshold unscientific. Environmental factors (such as temperature and humidity) are not factored into threshold calculations, leading to a high false alarm rate (e.g., high temperatures causing sensors to trigger falsely).

[0085] Therefore, a dynamic health threshold is set in this embodiment, and the dynamic health threshold is used as the basis for judging whether the health index meets the standard. Specifically, environmental data can be determined; based on the Bayesian optimization algorithm, and using the environmental data, the dynamic health threshold is determined. In other words, the "health threshold" in this embodiment can be adjusted in a timely manner according to the objective environment to adapt to the different working conditions of the equipment in different environments, so as to more objectively judge whether there is a fault or not, and shield the impact of environmental changes on the sensor data. For example, if the air conditioning load is high in summer, the temperature alarm threshold can be adjusted from 70°C to 75°C. This will not affect the normal operation of the equipment, and at the same time take into account the analysis and judgment of the fault.

[0086] By setting dynamic health thresholds, the rigidity of the fixed threshold alarm mechanism is solved.

[0087] Step 204: When the health index does not reach the dynamic health threshold, determine the fault type and fault probability based on the spatial characteristics and the temporal characteristics.

[0088] Step 205: Use the fault type and the fault probability as the prediction analysis result.

[0089] If the health index ranges from 0 to 100, with higher values indicating greater health, and the dynamic health threshold being 60, then if the actual health index falls below 60, it indicates a risk of failure. In other words, the health index has not reached the dynamic health threshold.

[0090] In this case, the fault type and the fault probability will be further determined based on the spatial characteristics and the temporal characteristics.

[0091] In this embodiment, the fault type can be first determined based on spatial characteristics. Spatial characteristics reflect specific "data values" and "data forms" in the sensor data. The health of the device and the causes of its "unhealthy" status are often reflected in this "data form." The analysis model not only quantifies this to produce a health index but also, by combining learning from historical data, conducts a detailed analysis of the spatial characteristics to determine the specific type of potential fault.

[0092] Furthermore, temporal features reflect the "time dependency" of sensor data (e.g., increased vibration for three consecutive days). This reflects the relationship between abnormal data and past events. This allows for further predictions of its relationship to future events (also through integration with historical data). In other words, it predicts the probability of a future failure and the time at which it might actually occur.

[0093] The above-mentioned fault types and fault probabilities serve as the prediction analysis results in this embodiment.

[0094] Step 206: Determine a maintenance work order based on the prediction analysis result.

[0095] Step 207: Determine maintenance resources and a work path of at least one candidate personnel, and determine the allocation strategy based on the maintenance resources and the work path.

[0096] After determining the predictive analysis results, this embodiment further processes them. Specifically, a maintenance work order is created based on the predictive analysis results and then assigned. For example, if the predictive analysis results indicate an 85% probability of failure for the main bearing of elevator A within three days, it can be repaired and inspected in advance. In other words, a maintenance work order can be created for this task.

[0097] When dispatching maintenance work orders, you can consider maintenance resources and work paths. Maintenance resources refer to the equipment, consumables, and other resources necessary to execute a maintenance work order. Work paths are the paths that different maintenance personnel must take (within the originally planned task path) to execute the maintenance work order. The length of these paths impacts overall execution efficiency.

[0098] Specifically, it can automatically match maintenance resources. For example, it can filter out maintenance personnel within 5 kilometers who hold elevator maintenance certificates and are currently unassigned. Then, it can use genetic algorithms to optimize the path to find the maintenance personnel with the shortest path and assign the maintenance work order to them.

[0099] It should also be noted that the system involved in this embodiment can further incorporate a closed-loop optimization mechanism. For example, after the owner reports a fault via the app, the reported result can be compared with the predictive analysis results to verify the accuracy of the analysis model's predictions. If the predictions are accurate, the model weighting is strengthened. After the repair is completed, the maintenance personnel fill in the actual cause of the fault, which can also verify the accuracy of the predictions.

[0100] like Figure 3 The figure shows a specific embodiment of the fault detection device in the property service of the present invention. The device described in this embodiment is used to perform Figures 1-2 The physical device of the method. Its technical solution is essentially consistent with the above embodiment, and the corresponding description in the above embodiment is also applicable to this embodiment. The device in this embodiment includes:

[0101] The data acquisition module 301 is used to collect sensor data in the target area using the Internet of Things sensor components and upload the sensor data to the analysis model.

[0102] The data analysis module 302 is configured to determine the health index corresponding to the sensor data using the analysis model.

[0103] The dynamic threshold module 303 is configured to determine a dynamic health threshold using the analysis model.

[0104] The prediction analysis module 304 is configured to determine a prediction analysis result based on the health index and the dynamic health threshold.

[0105] In addition Figure 3 Based on the embodiment shown, preferably, the present invention further includes:

[0106] The data acquisition module 301 includes:

[0107] The sensing unit 311 is used to collect sensing data in the target area using the Internet of Things sensing components.

[0108] The preprocessing unit 312 is configured to preprocess the sensor data using an edge computing node.

[0109] The Internet of Things communication unit 313 is used to upload the pre-processed sensor data to the analysis model.

[0110] The method further includes a training module 305 for determining an original neural network; performing data training on the original neural network using historical data; and using the trained original neural network as the analysis model.

[0111] The data analysis module 302 includes:

[0112] The feature extraction unit 321 is configured to extract the spatial features and temporal features of the sensing data using the analysis model.

[0113] The health index determining unit 322 is configured to determine the health index according to the spatial feature and the temporal feature.

[0114] The dynamic threshold module 303 includes:

[0115] The environment data determining unit 331 is configured to determine environment data.

[0116] The threshold determination unit 332 is configured to determine the dynamic health threshold based on a Bayesian optimization algorithm and using the environmental data.

[0117] The prediction analysis module 304 includes:

[0118] The fault determination unit 341 is configured to determine a fault type and a fault probability based on the spatial feature and the temporal feature when the health index does not reach the dynamic health threshold.

[0119] The prediction unit 342 is configured to use the fault type and the fault probability as the prediction analysis result.

[0120] The system further includes a work order module 306 for determining a maintenance work order according to the prediction analysis result; determining maintenance resources and a work path of at least one candidate personnel; and determining the allocation strategy according to the maintenance resources and the work path.

[0121] Figure 4 : This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include internal memory, such as high-speed random access memory (RAM), and may also include non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0122] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0123] Memory is used to store execution instructions. Specifically, execution instructions are computer programs that can be executed. Memory can include internal memory and non-volatile memory, and provides execution instructions and data to the processor.

[0124] In one possible implementation, a processor reads corresponding execution instructions from a non-volatile memory into a memory and then executes them. Alternatively, the processor can obtain corresponding execution instructions from other devices to logically form a fault detection device for property management services. The processor executes the execution instructions stored in the memory to implement the fault detection method for property management services provided in any embodiment of the present invention.

[0125] The present invention Figure 3 The method performed by the fault detection device in the property service provided in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor or the instructions in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The various methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0126] The steps of the method disclosed in conjunction with the embodiments of the present invention can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software modules can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the method described above.

[0127] The embodiment of the present invention further provides a readable medium, which stores an execution instruction. When the stored execution instruction is executed by the processor of the electronic device, the electronic device can execute the fault detection method in the property service provided in any embodiment of the present invention, and is specifically used to execute the following Figure 1 or Figure 2 The method shown.

[0128] The electronic device described in each of the aforementioned embodiments may be a computer.

[0129] Those skilled in the art will appreciate that the embodiments of the present invention may be provided as methods or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or a combination of software and hardware.

[0130] The various embodiments of the present invention are described in a progressive manner. Similar portions between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device embodiments are generally similar to the method embodiments, so their description is relatively simple. For relevant portions, refer to the description of the method embodiments.

[0131] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0132] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. A fault detection method in property services, characterized in that: include: Using IoT sensor components to collect sensor data within the target area and upload the sensor data to the analysis model; Determining a health index corresponding to the sensor data using the analysis model; determining a dynamic health threshold using the analytical model; A prediction analysis result is determined based on the health index and the dynamic health threshold.

2. The method according to claim 1, characterized in that The uploading of the sensor data to the analysis model comprises: The sensor data is preprocessed using an edge computing node, and the preprocessed sensor data is uploaded to the analysis model.

3. The method according to claim 1, characterized in that Also includes: Determine the original neural network; Performing data training on the original neural network using historical data; The trained original neural network is used as the analysis model.

4. The method according to claim 1, characterized in that Determining the health index corresponding to the sensor data by using the analysis model includes: extracting spatial and temporal features of the sensor data using the analysis model; The health index is determined according to the spatial characteristics and the temporal characteristics.

5. The method according to claim 4, characterized in that: Determining the dynamic health threshold using the analysis model includes: Determine environmental data; The dynamic health threshold is determined based on a Bayesian optimization algorithm and using the environmental data.

6. The method according to claim 5, characterized in that Determining the prediction analysis result according to the health index and the dynamic health threshold includes: When the health index does not reach the dynamic health threshold, determining a fault type and a fault probability based on the spatial characteristics and the temporal characteristics; The fault type and the fault probability are used as the prediction analysis result.

7. The method according to any one of claims 1 to 6, characterized in that: Also includes: determining a maintenance work order based on the predictive analysis results; Determine the work path of the maintenance resource and at least one backup personnel; The allocation strategy is determined according to the maintenance resources and the work path.

8. A fault detection device in property services, characterized in that: include: A data acquisition module, configured to collect sensor data within a target area using IoT sensor components and upload the sensor data to an analysis model; a data analysis module, configured to determine a health index corresponding to the sensor data using the analysis model; a dynamic threshold module, configured to determine a dynamic health threshold using the analysis model; The prediction analysis module is used to determine the prediction analysis result based on the health index and the dynamic health threshold.

9. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the fault detection method in property services described in any one of claims 1 to 7.

10. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the fault detection method in property services described in any one of claims 1 to 7.

Images