A property management method for smart community

Abstract

The present disclosure relates to a property management method for a smart community. The method includes: performing signal segmentation on the acquired water flow pattern ultrasonic detection signal, and then performing feature extraction through a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors; calculating the anchor center of the sequence of water flow pattern ultrasonic detection waveform feature vectors, and calculating the hash similarity between the sequence of water flow pattern ultrasonic detection waveform feature vectors and the anchor center, so as to determine whether there is a leak based on the hash similarity, and determine whether to generate a leak warning prompt. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

Description

Technical Field

[0001] The present disclosure relates to the technical field of smart communities, and in particular, to a property management method for smart communities. Background Art

[0002] Smart community property management refers to a series of activities to maintain, manage and provide services for buildings, facilities, environment, etc. in smart communities. Smart community property management plays a vital role in today's society. It is not only related to the maintenance of the community environment and the improvement of residents' quality of life, but also directly affects the safety and sustainable development of the community. However, traditional smart community property management methods have many defects, among which water resource management is particularly prominent.

[0003] In traditional smart communities, water resource management usually relies on manual meter reading to monitor and manage water resource usage. Manual meter reading requires a lot of manpower and material resources, is inefficient, and is prone to human errors and omissions. It also fails to detect water pipe leaks in a timely manner, resulting in water pipe leaks not being dealt with in a timely manner, causing a waste of water resources.

[0004] Therefore, an optimized property management solution for smart communities is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present disclosure provides a property management method for a smart community, the method comprising:

[0007] Acquire ultrasonic detection signals of water flow patterns collected by smart water meters;

[0008] Segmenting the water flow mode ultrasonic detection signal into signal segments to obtain a sequence of water flow mode ultrasonic detection signal segments;

[0009] Using a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model, feature extraction is performed on each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors;

[0010] Calculating the anchor center of the sequence of the water flow pattern ultrasonic detection waveform feature vectors, the anchor center is the position mean vector of the sequence of the water flow pattern ultrasonic detection waveform feature vectors;

[0011] Calculating the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and performing temporal stability characterization of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature;

[0012] Based on the hash similarity, it is determined whether there is a leak and whether to generate a leak warning prompt.

[0013] Optionally, the deep neural network model is a convolutional neural network model.

[0014] Optionally, a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model is used to perform feature extraction on each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors, including: using each layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model to perform convolution processing, mean pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer so that the output of the last layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is the sequence of water flow pattern ultrasonic detection waveform feature vectors, wherein the input of the first layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments.

[0015] Optionally, the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center is calculated, and the temporal stability of the water flow pattern is characterized based on the hash similarity to obtain a water flow pattern temporal stability characterization feature, including: calculating the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector relative to the anchor center to obtain a water flow pattern temporal stability characterization feature vector composed of multiple hash similarities as the water flow pattern temporal stability characterization feature.

[0016] Optionally, based on the hash similarity, determining whether there is a leak and determining whether to generate a leak warning prompt includes: passing the water flow pattern temporal stability characterization feature vector through a classifier-based leak detector to obtain a detection result, and the detection result is used to indicate whether there is a leak; in response to the detection result being that there is a leak, generating a leak warning prompt.

[0017] Optionally, a training step is also included: used to train the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the classifier-based leakage detector.

[0018] Optionally, the training step includes: obtaining a training water flow pattern ultrasonic detection signal collected by a smart water meter, and a true value of whether the smart water meter has a leak; dividing the training water flow pattern ultrasonic detection signal into signal segments to obtain a sequence of training water flow pattern ultrasonic detection signal segments; performing feature extraction on each training water flow pattern ultrasonic detection signal segment in the sequence of training water flow pattern ultrasonic detection signal segments through the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model to obtain a sequence of training water flow pattern ultrasonic detection waveform feature vectors; calculating a training anchor center of the sequence of training water flow pattern ultrasonic detection waveform feature vectors, wherein the training anchor center is the training water flow pattern ultrasonic detection signal. The method comprises the following steps: detecting a positional mean vector of a sequence of waveform feature vectors; calculating the training hash similarity of each training water flow pattern ultrasonic detection waveform feature vector in the sequence of training water flow pattern ultrasonic detection waveform feature vectors relative to the training anchor center to obtain a training water flow pattern temporal stability characterization feature vector composed of multiple training hash similarities; passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leakage detector to obtain a classification loss function value; training the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the leakage detector based on the classifier based on the classification loss function value, and optimizing the training water flow pattern temporal stability characterization feature vector in each iteration.

[0019] Optionally, passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leakage detector to obtain a classification loss function value, including: using the classifier-based leakage detector to process the training water flow pattern temporal stability characterization feature vector using the following training classification formula to generate a training classification result, wherein the training classification formula is: ,in represents the characteristic vector representing the temporal stability of the training water flow pattern, to is the weight matrix, to represents a bias matrix; and calculating a cross entropy value between the training classification result and the true value as the classification loss function value.

[0020] The above technical solution is adopted to segment the acquired water flow pattern ultrasonic detection signal into signal segments, and then extract features through a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors; the anchor center of the sequence of water flow pattern ultrasonic detection waveform feature vectors is calculated, and the hash similarity between the sequence of water flow pattern ultrasonic detection waveform feature vectors and the anchor center is calculated to determine whether there is a leak based on the hash similarity, and determine whether to generate a leak warning prompt. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

[0021] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0023] Figure 1 It is a flow chart of a property management method for a smart community according to an exemplary embodiment.

[0024] Figure 2 It is a block diagram of a property management system for a smart community according to an exemplary embodiment.

[0025] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0026] Figure 4 It is an application scenario diagram of a property management method for a smart community according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0028] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0029] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0030] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0031] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0032] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0033] The modernization and scientificization of smart community property management is a key factor in promoting the sustainable development of the community. In terms of water resource management, smart community property management can achieve efficient use and rational allocation of water resources by introducing advanced technologies and management concepts, thereby solving the problems existing in traditional management methods.

[0034] First of all, smart community property management can monitor water usage in real time by installing smart water meters and monitoring systems. These systems can collect and analyze data to help property managers detect abnormal water use behavior in a timely manner, thereby effectively preventing waste of water resources and illegal water use.

[0035] Secondly, smart community property management can adopt water-saving technologies and equipment, such as rainwater collection systems and gray water recycling systems, to increase the reuse rate of water resources. The rainwater collection system can collect rainwater in the rainy season and use it for non-drinking purposes such as green irrigation and cleaning after treatment, reducing the use of tap water. The gray water recycling system can collect gray water from domestic water, and after filtering and treatment, it can be used again for flushing toilets, etc., further saving water resources.

[0036] In addition, smart community property management can also improve residents' awareness of water conservation through publicity, education and incentive mechanisms. The property management department can regularly hold lectures on water conservation knowledge, and let residents understand the importance of water conservation and water conservation skills through publicity materials and on-site demonstrations. At the same time, a water conservation reward mechanism can be set up to give certain material or honorary rewards to residents or families with outstanding water conservation performance to stimulate residents' enthusiasm for water conservation.

[0037] Finally, smart community property management should strengthen communication and collaboration with relevant departments, and actively participate in the formulation and implementation of water resource management and protection policies. The property management department can cooperate with environmental protection, water conservancy and other departments to jointly carry out water resource surveys, water quality monitoring and other work to ensure the rational use and effective protection of community water resources.

[0038] In order to solve the problems existing in water resource management in traditional smart communities, modern smart communities have begun to use digital and intelligent technical means to monitor and manage water resources. These technical means include but are not limited to the Internet of Things (IoT), cloud computing, big data analysis, artificial intelligence (AI), etc., which can significantly improve the efficiency and accuracy of water resource management while reducing the investment of manpower and material resources.

[0039] The present disclosure provides a property management method for a smart community, which divides the acquired water flow pattern ultrasonic detection signal into signal segments, and then extracts features through a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors; calculates the anchor center of the sequence of water flow pattern ultrasonic detection waveform feature vectors, and calculates the hash similarity between the sequence of water flow pattern ultrasonic detection waveform feature vectors and the anchor center, so as to determine whether there is a leak based on the hash similarity, and determine whether to generate a leak warning prompt. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

[0040] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0041] It should be understood that the ultrasonic sensor is a transducer. By installing the ultrasonic sensor in the smart water meter on the water pipe and emitting ultrasonic pulses into the water to collect ultrasonic detection signals, it can measure the pattern of water flow and abnormalities or leakage. Specifically, when there is no leak in the water, the ultrasonic pulse will propagate at a constant speed and pattern. However, when a leak occurs, the water flow pattern will change, which will cause the propagation speed and pattern of the ultrasonic pulse to change. The ultrasonic sensor will detect these changes and trigger an alarm, so as to detect the leakage problem in time and avoid the waste of water resources, thereby improving the property management efficiency and quality of life of residents in the smart community.

[0042] Based on this, in the technical solution of this application, a property management method for smart communities is proposed, which can monitor and collect water flow pattern ultrasonic detection signals in real time through smart water meters, and use artificial intelligence-based signal processing and analysis algorithms at the back end to analyze the water flow pattern ultrasonic detection signals, so as to detect whether there is a water resource leakage problem, and issue a leakage alarm in time when the leakage problem occurs. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

[0043] Figure 1 is a flow chart of a property management method for a smart community according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0044] Step 101, obtaining a water flow pattern ultrasonic detection signal collected by a smart water meter;

[0045] Step 102, dividing the water flow mode ultrasonic detection signal into signal segments to obtain a sequence of water flow mode ultrasonic detection signal segments;

[0046] Step 103: extracting features of each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments by using a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors;

[0047] Step 104, calculating the anchor center of the sequence of the water flow pattern ultrasonic detection waveform feature vectors, wherein the anchor center is the position mean vector of the sequence of the water flow pattern ultrasonic detection waveform feature vectors;

[0048] Step 105, calculating the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and performing temporal stability characterization of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature;

[0049] Step 106: Based on the hash similarity, determine whether there is a leak and whether to generate a leak warning prompt.

[0050] Among them, the deep neural network model is a convolutional neural network model.

[0051] Specifically, in the technical solution of the present application, first, the water flow pattern ultrasonic detection signal collected by the smart water meter is obtained. Next, considering that there is a large amount of information in the water flow pattern ultrasonic detection signal, it not only contains characteristic information about the water flow pattern, such as the characteristics and change law of the water flow, but also contains some interference features that are irrelevant to the water flow pattern detection. Therefore, in order to better analyze and identify the water flow pattern, thereby realizing the intelligent and precise management of water resources, in the technical solution of the present application, it is necessary to segment the water flow pattern ultrasonic detection signal into signal segments to obtain a sequence of water flow pattern ultrasonic detection signal segments. By segmenting the continuous ultrasonic signal into segments, the water flow pattern ultrasonic detection signal can be made easier to process and analyze. Each segment represents the water flow pattern information within a specific time period. By analyzing these segments, the characteristics and change law of the water flow can be more accurately understood. In addition, segmenting the signal segments helps to extract the characteristics of the water flow pattern. By extracting the features of each segment, key waveform semantic feature information, such as amplitude, frequency, waveform shape, etc., can be obtained. These waveform semantic features are very important for identifying different water flow patterns and detecting abnormal conditions.

[0052] Then, each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments is subjected to feature mining in a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model, so as to extract semantic feature information of the water flow pattern ultrasonic detection waveform in different local time periods in each water flow pattern ultrasonic detection signal segment, thereby obtaining a sequence of water flow pattern ultrasonic detection waveform feature vectors. In this way, the features of the water flow pattern in different local time periods can be better represented, including information such as amplitude, frequency, and waveform shape, and these features can be used for subsequent pattern recognition and classification tasks.

[0053] In one embodiment of the present disclosure, a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model is used to perform feature extraction on each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors, including: using each layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model to perform convolution processing, mean pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer so that the output of the last layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is the sequence of water flow pattern ultrasonic detection waveform feature vectors, wherein the input of the first layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments.

[0054] Furthermore, it is considered that the water flow pattern characteristics of each local time period may change when the water flow pattern characteristics are actually monitored and analyzed, and there may also be some change patterns that deviate from the overall characteristics of the water flow pattern. That is to say, when there is no leakage in the water, the ultrasonic pulse will propagate at a constant speed and mode. However, when a leak occurs, the water flow pattern will change, which will cause the propagation speed and mode of the ultrasonic pulse to change. Therefore, in order to be able to characterize the overall characteristics of the water flow pattern in each local time period, so as to more comprehensively describe the timing characteristics and stability of the water flow pattern, in the technical solution of the present application, the anchor center of the sequence of the water flow pattern ultrasonic detection waveform feature vector is further calculated, and the anchor center is the position mean vector of the sequence of the water flow pattern ultrasonic detection waveform feature vector. In addition, the hash similarity of each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector is calculated relative to the anchor center to obtain a water flow pattern timing stability characterization feature vector composed of multiple hash similarities. By calculating the hash similarity of each water flow pattern ultrasonic detection waveform feature vector relative to the anchor center, the similarity between the semantic features of the water flow pattern ultrasonic detection waveform in each local time period and the overall water flow pattern features can be compared. This can be used to reflect through relative analysis whether the water flow pattern in each local time period excessively deviates from the overall water flow pattern change, thereby better evaluating the temporal stability of the water flow pattern and discovering abnormal water flow patterns and leakage problems.

[0055] In one embodiment of the present disclosure, the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center is calculated, and the temporal stability of the water flow pattern is characterized based on the hash similarity to obtain a water flow pattern temporal stability characterization feature, including: calculating the hash similarity of each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector relative to the anchor center to obtain a water flow pattern temporal stability characterization feature vector composed of multiple hash similarities as the water flow pattern temporal stability characterization feature.

[0056] Then, the characteristic vector representing the temporal stability of the water flow pattern is passed through a leak detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether there is a leak. In other words, the characteristic information representing the temporal stability of the water flow pattern is used for classification processing to detect water flow leakage, and then, in response to the detection result that there is a leak, a leak warning prompt is generated. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

[0057] In one embodiment of the present disclosure, based on the hash similarity, determining whether there is a leak and determining whether to generate a leak warning prompt includes: passing the water flow pattern temporal stability characterization feature vector through a classifier-based leak detector to obtain a detection result, and the detection result is used to indicate whether there is a leak; in response to the detection result being that there is a leak, generating a leak warning prompt.

[0058] In the technical scheme described above, each water flow pattern ultrasonic detection waveform feature vector in the sequence of water flow pattern ultrasonic detection waveform feature vectors respectively expresses the signal waveform image semantic features of the water flow pattern ultrasonic detection signal in the local time domain under the global time domain. In this way, the anchor center can express the global homogenized correlation features of the image semantics between the local time domains relative to the global time domain, so that the water flow pattern temporal stability characterization feature vector composed of multiple hash similarities expresses the local time domain-global time domain image semantic similarity stable mapping features.

[0059] However, considering the non-uniformity of the source image semantic mapping between the local time domain and the global time domain of the water flow pattern ultrasonic detection signal, the water flow pattern temporal stability characterization feature vector composed of multiple hash similarities will also have local distribution non-uniformity based on the eigenvalue granularity due to the difference in feature representation in the local time domain corresponding to each water flow pattern ultrasonic detection waveform feature vector. That is, the water flow pattern temporal stability characterization feature vector has the illusion of local-global non-uniform feature distribution mapping, thereby affecting the convergence effect of the water flow pattern temporal stability characterization feature vector to the class probability density space when classified by the classifier.

[0060] Accordingly, in the technical solution of the present application, the characteristic vector representing the temporal stability of the water flow pattern is passed through a leak detector based on a classifier to obtain a detection result, including:

[0061] Obtaining the characteristic vector representing the temporal stability of the water flow pattern and inputting it into the classifier-based leakage detector to obtain a leakage probability value corresponding to the presence of leakage;

[0062] Probabilistically converting each eigenvalue of the water flow pattern temporal stability characterization eigenvalue vector based on a probability activation function, such as a sigmoid function, to obtain a probabilistic water flow pattern temporal stability characterization eigenvalue vector;

[0063] Determine a cognitive symbol value for characterizing the temporal stability of the water flow pattern based on a comparison between each eigenvalue of the probabilistic water flow pattern temporal stability characterizing eigenvector and the leakage probability value, wherein the cognitive symbol value for characterizing the temporal stability of the water flow pattern is equal to one, zero and negative one respectively in response to the eigenvalue of the probabilistic water flow pattern temporal stability characterizing eigenvector being greater than, equal to and less than the leakage probability value;

[0064] Calculating the mean of all eigenvalues ​​of the eigenvector representing the temporal stability of the probabilistic water flow pattern to obtain an overall phase shift value representing the temporal stability of the water flow pattern;

[0065] Multiply each eigenvalue of the probabilistic water flow pattern temporal stability characterization eigenvector by the water flow pattern temporal stability characterization cognitive symbol value and the water flow pattern temporal stability characterization overall phase shift value to obtain a probabilistic water flow pattern temporal stability characterization cognitive eigenvector and a probabilistic water flow pattern temporal stability characterization phase shift eigenvector;

[0066] The probabilistic water flow pattern temporal stability characterization cognitive feature vector and the probabilistic water flow pattern temporal stability characterization phase shift feature vector are respectively multiplied by a first weight and a second weight as hyperparameters, and then the difference is calculated to obtain a probabilistic water flow pattern temporal stability characterization differential feature vector;

[0067] Taking the absolute value of each eigenvalue of the probabilistic water flow pattern temporal stability characterization differential eigenvector to obtain an optimized water flow pattern temporal stability characterization eigenvector;

[0068] The optimized water flow pattern temporal stability characterization feature vector is input into a classifier-based leakage detector to obtain the detection result.

[0069] Specifically, the optimization steps are expressed as:

[0070]

[0071] in, is the first eigenvector representing the temporal stability of the probabilistic flow pattern. The eigenvalues ​​at the positions, is the leakage probability value, is the cognitive symbol value representing the temporal stability of the water flow pattern, and denote the first weight and the second weight as hyperparameters, respectively. is the mean of all eigenvalues ​​of the eigenvector representing the temporal stability of the probabilistic water flow pattern, that is, the overall phase shift value representing the temporal stability of the water flow pattern, is the first eigenvector characterizing the temporal stability of the optimized water flow pattern. The feature value at each position.

[0072] Therefore, in the above-mentioned preferred embodiment of the present application, the class-cognitive phase shift response of the feature vector characterizing the temporal stability of the water pattern is obtained by comparing the probabilistic amplitude of the eigenvalue of the feature vector characterizing the temporal stability of the water pattern with the conforming probability representation, and the phase shift response of the eigenvalue of the feature vector characterizing the temporal stability of the water pattern relative to the class probability representation of the feature as a whole is subjected to a feature distribution sequence invariance transformation based on the class probability differential distribution expansion, so as to realize the causal constraint of the posterior class probability of the feature vector characterizing the temporal stability of the water pattern on its prior feature distribution representation, so as to improve the convergence effect of the feature vector characterizing the temporal stability of the water pattern to the class probability density space when the feature vector is classified by the classifier, thereby improving the accuracy of the classification result obtained by inputting the feature vector characterizing the temporal stability of the water pattern into the classifier.

[0073] In one embodiment of the present disclosure, the property management method for a smart community further includes a training step: for training the water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model and the leak detector based on a classifier. The training step includes: obtaining a training water flow pattern ultrasonic detection signal collected by a smart water meter, and a true value of whether there is a leak in the smart water meter; segmenting the training water flow pattern ultrasonic detection signal into signal segments to obtain a sequence of training water flow pattern ultrasonic detection signal segments; performing feature extraction on each training water flow pattern ultrasonic detection signal segment in the sequence of training water flow pattern ultrasonic detection signal segments by the water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of training water flow pattern ultrasonic detection waveform feature vectors; calculating a training anchor center of the sequence of training water flow pattern ultrasonic detection waveform feature vectors, wherein the training anchor center is the training water flow pattern ultrasonic detection signal segment. The method comprises the following steps: obtaining a water flow pattern ultrasonic detection waveform feature vector having a positional mean vector of a sequence of waveform feature vectors; calculating the training hash similarity of each training water flow pattern ultrasonic detection waveform feature vector in the sequence of training water flow pattern ultrasonic detection waveform feature vectors relative to the training anchor center to obtain a training water flow pattern temporal stability characterization feature vector composed of multiple training hash similarities; passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leakage detector to obtain a classification loss function value; training the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the leakage detector based on the classifier based on the classification loss function value, and optimizing the training water flow pattern temporal stability characterization feature vector in each iteration.

[0074] Further, the training water flow pattern temporal stability characterization feature vector is passed through the classifier-based leakage detector to obtain a classification loss function value, including: using the classifier-based leakage detector to process the training water flow pattern temporal stability characterization feature vector using the following training classification formula to generate a training classification result, wherein the training classification formula is: ,in represents the characteristic vector representing the temporal stability of the training water flow pattern, to is the weight matrix, to represents a bias matrix; and calculating a cross entropy value between the training classification result and the true value as the classification loss function value.

[0075] In one embodiment of the present application, each time the training water flow pattern temporal stability characterization feature vector is classified and iterated by the classifier, the training water flow pattern temporal stability characterization feature vector is optimized, which is expressed as follows: in each iteration, the training water flow pattern temporal stability characterization feature vector is optimized to obtain the optimized training water flow pattern temporal stability characterization feature vector; wherein the optimization formula is:

[0076]

[0077] in, and They are the characteristic vectors representing the temporal stability of the training flow pattern No. and The eigenvalues ​​of the positions, is the characteristic vector representing the temporal stability of the training flow pattern, is the characteristic vector representing the temporal stability of the optimized training flow pattern, represents matrix multiplication, Represents positional addition.

[0078] That is, by introducing the training flow pattern temporal stability characterization feature vector The local statistical information distribution is used as an external information source to enhance the retrieval of the feature vector, so as to avoid the local overflow information distribution causing the temporal stability of the training water flow pattern to characterize the feature vector based on the local statistical dense information structure. The non-uniform feature distribution mapping illusion is used to obtain the characteristic vector representing the temporal stability of the training water flow pattern Information credible response reasoning based on the preservation of the local distribution group dimension corresponding to the ultrasonic detection waveform feature vector of each water flow pattern, that is, obtaining the temporal stability characterization feature vector of the training water flow pattern Based on the credible distribution response in the probability density space of the non-uniform local feature distribution, the convergence effect of the probability density space is improved to improve the training speed and the accuracy of the training results. In this way, water resource leakage detection and management in smart communities can be more accurately carried out to detect leakage problems in time, thereby avoiding the waste of water resources and improving the property management efficiency and quality of life of residents in smart communities.

[0079] In summary, the above scheme is adopted to collect ultrasonic detection signals of water flow patterns through real-time monitoring of smart water meters, and to analyze the ultrasonic detection signals of water flow patterns using signal processing and analysis algorithms based on artificial intelligence at the back end, so as to detect whether there is a water resource leakage problem and issue a leakage alarm in time when a leakage problem occurs. In this way, water resource leakage detection and management in smart communities can be carried out intelligently, property management efficiency and residents' quality of life can be improved, and a safer and more convenient living environment can be provided for residents.

[0080] Figure 2 FIG. 1 is a block diagram of a property management system for a smart community according to an exemplary embodiment. Figure 2 As shown, the system 200 includes:

[0081] The water flow pattern ultrasonic detection signal acquisition module 201 is used to acquire the water flow pattern ultrasonic detection signal collected by the smart water meter;

[0082] A signal segmentation module 202, configured to segment the water flow mode ultrasonic detection signal into signal segments to obtain a sequence of water flow mode ultrasonic detection signal segments;

[0083] A water flow pattern ultrasonic signal waveform feature extraction module 203 is used to extract features of each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments through a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors;

[0084] An anchor center calculation module 204, used to calculate the anchor center of the sequence of water flow pattern ultrasonic detection waveform feature vectors, the anchor center is the position mean vector of the sequence of water flow pattern ultrasonic detection waveform feature vectors;

[0085] A hash similarity calculation module 205 is used to calculate the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and to characterize the temporal stability of the water flow pattern based on the hash similarity to obtain the water flow pattern temporal stability characterization feature;

[0086] The leakage determination module 206 is used to determine whether there is a leakage based on the hash similarity and determine whether to generate a leakage warning prompt.

[0087] Reference below Figure 3, which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0088] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0089] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0090] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0091] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0092] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0093] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0094] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0095] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0096] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device".

[0097] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0098] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0099] According to one or more embodiments of the present disclosure, Example 1 provides a property management method for a smart community, the method comprising:

[0100] Acquire ultrasonic detection signals of water flow patterns collected by smart water meters;

[0101] Segmenting the water flow mode ultrasonic detection signal into signal segments to obtain a sequence of water flow mode ultrasonic detection signal segments;

[0102] Using a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model, feature extraction is performed on each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors;

[0103] Calculating the anchor center of the sequence of the water flow pattern ultrasonic detection waveform feature vectors, the anchor center is the position mean vector of the sequence of the water flow pattern ultrasonic detection waveform feature vectors;

[0104] Calculating the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and performing temporal stability characterization of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature;

[0105] Based on the hash similarity, it is determined whether there is a leak and whether to generate a leak warning prompt.

[0106] According to one or more embodiments of the present disclosure, Example 2 provides the method of Example 1, wherein the deep neural network model is a convolutional neural network model.

[0107] According to one or more embodiments of the present disclosure, Example 3 provides the method of Example 2, wherein a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model is used to extract features of each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors, including:

[0108] Each layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is used to perform convolution processing, mean pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer so that the output of the last layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is a sequence of the water flow pattern ultrasonic detection waveform feature vectors, wherein the input of the first layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments.

[0109] According to one or more embodiments of the present disclosure, Example 4 provides the method of Example 3, calculating the hash similarity of each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and performing temporal stability characterization of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature, including: calculating the hash similarity of each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector relative to the anchor center to obtain a water flow pattern temporal stability characterization feature vector composed of multiple hash similarities as the water flow pattern temporal stability characterization feature.

[0110] According to one or more embodiments of the present disclosure, Example 5 provides the method of Example 4, and determines whether there is a leak based on the hash similarity, and determines whether to generate a leak warning prompt, including:

[0111] Passing the water flow pattern temporal stability characterization feature vector through a classifier-based leak detector to obtain a detection result, wherein the detection result is used to indicate whether there is a leak;

[0112] In response to the detection result indicating leakage, a leakage warning prompt is generated.

[0113] According to one or more embodiments of the present disclosure, Example 6 provides the method of Example 5, further comprising a training step: for training the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the classifier-based leakage detector.

[0114] According to one or more embodiments of the present disclosure, Example 7 provides the method of Example 6, wherein the training step includes:

[0115] Obtaining the ultrasonic detection signal of the training water flow pattern collected by the smart water meter, and the real value of whether the smart water meter has a leak;

[0116] Segmenting the training water flow pattern ultrasonic detection signal into signal segments to obtain a sequence of training water flow pattern ultrasonic detection signal segments;

[0117] The water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model performs feature extraction on each training water flow pattern ultrasonic detection signal segment in the sequence of training water flow pattern ultrasonic detection signal segments to obtain a sequence of training water flow pattern ultrasonic detection waveform feature vectors;

[0118] Calculating a training anchor center of the sequence of training water flow pattern ultrasonic detection waveform feature vectors, wherein the training anchor center is a position mean vector of the sequence of training water flow pattern ultrasonic detection waveform feature vectors;

[0119] Calculating the training hash similarity of each training water flow pattern ultrasonic detection waveform feature vector in the sequence of the training water flow pattern ultrasonic detection waveform feature vector relative to the training anchor center to obtain a training water flow pattern temporal stability characterization feature vector composed of multiple training hash similarities;

[0120] Passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leak detector to obtain a classification loss function value;

[0121] The water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the classifier-based leakage detector are trained based on the classification loss function value, and in each iteration, the feature vector characterizing the temporal stability of the training water flow pattern is optimized.

[0122] According to one or more embodiments of the present disclosure, Example 8 provides the method of Example 7, wherein the training water flow pattern temporal stability characterization feature vector is passed through the classifier-based leak detector to obtain a classification loss function value, including:

[0123] The classifier-based leak detector is used to process the training water flow pattern temporal stability characterization feature vector using the following training classification formula to generate a training classification result, wherein the training classification formula is: ,in represents the characteristic vector representing the temporal stability of the training water flow pattern, to is the weight matrix, to represents the bias matrix; and

[0124] The cross entropy value between the training classification result and the true value is calculated as the classification loss function value.

[0125] Figure 4 FIG. 1 is an application scenario diagram of a property management method for a smart community according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, the water flow pattern ultrasonic detection signal collected by the smart water meter is obtained (for example, Figure 4 Then, the acquired water flow pattern ultrasonic detection signal is input to a server (for example, Figure 4 In S) shown in the figure, the server is capable of processing the water flow pattern ultrasonic detection signal based on the property management algorithm for the smart community to determine whether there is a leak and whether to generate a leak warning prompt.

[0126] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0127] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0128] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A property management method for a smart community, characterized in that: include: Acquire ultrasonic detection signals of water flow patterns collected by smart water meters; Segmenting the water flow mode ultrasonic detection signal into signal segments to obtain a sequence of water flow mode ultrasonic detection signal segments; Using a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model, feature extraction is performed on each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors; Calculating the anchor center of the sequence of the water flow pattern ultrasonic detection waveform feature vectors, the anchor center is the position mean vector of the sequence of the water flow pattern ultrasonic detection waveform feature vectors; Calculating the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and performing temporal stability characterization of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature; Based on the hash similarity, determining whether there is a leak and determining whether to generate a leak warning prompt; Based on the hash similarity, determining whether there is a leak and determining whether to generate a leak warning prompt includes: Passing the water flow pattern temporal stability characterization feature vector through a classifier-based leak detector to obtain a detection result, wherein the detection result is used to indicate whether there is a leak; In response to the detection result indicating leakage, generating a leakage warning prompt; Calculate the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center, and characterize the temporal stability of the water flow pattern based on the hash similarity to obtain a water flow pattern temporal stability characterization feature, including: calculate the hash similarity between each water flow pattern ultrasonic detection waveform feature vector in the sequence of the water flow pattern ultrasonic detection waveform feature vector and the anchor center to obtain a water flow pattern temporal stability characterization feature vector composed of multiple hash similarities as the water flow pattern temporal stability characterization feature; The water flow pattern temporal stability characterization feature vector is passed through a classifier-based leak detector to obtain a detection result, including: Obtaining the characteristic vector representing the temporal stability of the water flow pattern and inputting it into the classifier-based leakage detector to obtain a leakage probability value corresponding to the presence of leakage; Probabilistically converting each eigenvalue of the water flow pattern temporal stability characterization eigenvalue vector based on the probability activation function sigmoid function to obtain a probabilistic water flow pattern temporal stability characterization eigenvalue vector; Determine a cognitive symbol value for characterizing the temporal stability of the water flow pattern based on a comparison between each eigenvalue of the probabilistic water flow pattern temporal stability characterizing eigenvector and the leakage probability value, wherein the cognitive symbol value for characterizing the temporal stability of the water flow pattern is equal to one, zero and negative one respectively in response to the eigenvalue of the probabilistic water flow pattern temporal stability characterizing eigenvector being greater than, equal to and less than the leakage probability value; Calculating the mean of all eigenvalues ​​of the eigenvector representing the temporal stability of the probabilistic water flow pattern to obtain an overall phase shift value representing the temporal stability of the water flow pattern; Multiply each eigenvalue of the probabilistic water flow pattern temporal stability characterization eigenvector by the water flow pattern temporal stability characterization cognitive symbol value and the water flow pattern temporal stability characterization overall phase shift value to obtain a probabilistic water flow pattern temporal stability characterization cognitive eigenvector and a probabilistic water flow pattern temporal stability characterization phase shift eigenvector; The probabilistic water flow pattern temporal stability characterization cognitive feature vector and the probabilistic water flow pattern temporal stability characterization phase shift feature vector are respectively multiplied by a first weight and a second weight as hyperparameters, and then the difference is calculated to obtain a probabilistic water flow pattern temporal stability characterization differential feature vector; Taking the absolute value of each eigenvalue of the probabilistic water flow pattern temporal stability characterization differential eigenvector to obtain an optimized water flow pattern temporal stability characterization eigenvector; Inputting the optimized water flow pattern temporal stability characterization feature vector into a classifier-based leak detector to obtain the detection result; The optimization steps are expressed as: ; ; in, is the first eigenvector representing the temporal stability of the probabilistic flow pattern. The eigenvalues ​​at the positions, is the leakage probability value, is the cognitive symbol value representing the temporal stability of the water flow pattern, and denote the first weight and the second weight as hyperparameters, respectively. is the mean of all eigenvalues ​​of the eigenvector representing the temporal stability of the probabilistic water flow pattern, that is, the overall phase shift value representing the temporal stability of the water flow pattern, is the first eigenvector characterizing the temporal stability of the optimized water flow pattern. The feature value at each position.

2. The property management method for a smart community according to claim 1, characterized in that: The deep neural network model is a convolutional neural network model.

3. The property management method for a smart community according to claim 2, characterized in that: The method comprises: extracting features of each water flow pattern ultrasonic detection signal segment in the sequence of water flow pattern ultrasonic detection signal segments by using a water flow pattern ultrasonic signal waveform feature extractor based on a deep neural network model to obtain a sequence of water flow pattern ultrasonic detection waveform feature vectors, including: Each layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is used to perform convolution processing, mean pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer so that the output of the last layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is a sequence of the water flow pattern ultrasonic detection waveform feature vectors, wherein the input of the first layer of the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model is each water flow pattern ultrasonic detection signal segment in the sequence of the water flow pattern ultrasonic detection signal segments.

4. The property management method for a smart community according to claim 3, characterized in that: It also includes a training step: used to train the water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the leakage detector based on the classifier.

5. The property management method for a smart community according to claim 4, characterized in that: The training step comprises: Obtaining the ultrasonic detection signal of the training water flow pattern collected by the smart water meter, and the real value of whether the smart water meter has a leak; Segmenting the training water flow pattern ultrasonic detection signal into signal segments to obtain a sequence of training water flow pattern ultrasonic detection signal segments; The water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model performs feature extraction on each training water flow pattern ultrasonic detection signal segment in the sequence of training water flow pattern ultrasonic detection signal segments to obtain a sequence of training water flow pattern ultrasonic detection waveform feature vectors; Calculating a training anchor center of the sequence of training water flow pattern ultrasonic detection waveform feature vectors, wherein the training anchor center is a position mean vector of the sequence of training water flow pattern ultrasonic detection waveform feature vectors; Calculating the training hash similarity of each training water flow pattern ultrasonic detection waveform feature vector in the sequence of the training water flow pattern ultrasonic detection waveform feature vector relative to the training anchor center to obtain a training water flow pattern temporal stability characterization feature vector composed of multiple training hash similarities; Passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leak detector to obtain a classification loss function value; The water flow pattern ultrasonic signal waveform feature extractor based on the deep neural network model and the classifier-based leakage detector are trained based on the classification loss function value, and in each iteration, the feature vector characterizing the temporal stability of the training water flow pattern is optimized.

6. The property management method for a smart community according to claim 5, characterized in that: Passing the training water flow pattern temporal stability characterization feature vector through the classifier-based leak detector to obtain a classification loss function value, including: The classifier-based leak detector is used to process the training water flow pattern temporal stability characterization feature vector using the following training classification formula to generate a training classification result, wherein the training classification formula is: ,in represents the characteristic vector representing the temporal stability of the training water flow pattern, to is the weight matrix, to represents the bias matrix; and The cross entropy value between the training classification result and the true value is calculated as the classification loss function value.

Images