Environment-friendly public toilet equipment state monitoring method, device and equipment

Abstract

The application provides an environmental public toilet equipment state monitoring method, device and equipment, and relates to the technical field of equipment state monitoring.In the application, first, multi-dimensional state time sequence data of a target environmental public toilet equipment is acquired;second, cross-dimension state time sequence semantic mining is performed on the multi-dimensional state time sequence data to form multi-dimensional state time sequence characteristics corresponding to the multi-dimensional state time sequence data;then, network state prediction is performed based on the multi-dimensional state time sequence characteristics to obtain a target network state prediction result.Based on the above, the problem that the reliability of equipment state monitoring is relatively low in the prior art can be improved.

Description

Technical Field

[0001] This application relates to the field of equipment status monitoring technology, and more specifically, to a method, device, and equipment for monitoring the status of environmentally friendly public toilet equipment. Background Technology

[0002] Smart and eco-friendly public toilets rely on various online devices to achieve their intelligent functions. Device offline status can lead to data loss and service interruptions. Traditional equipment management methods struggle to monitor the online status of devices in real time, making it difficult to promptly detect and address offline issues. With the development of IoT technology, building remote offline status monitoring and alarm systems has become possible, significantly improving the operational efficiency and reliability of public toilets. Existing smart public toilet equipment management systems mostly rely on simple, timed heartbeat detection for monitoring offline status, lacking analysis and diagnosis of the causes of offline issues. However, this approach can only be applied after a network failure occurs, making timely and effective maintenance difficult. Therefore, some conventional solutions determine potential network failures in eco-friendly public toilet equipment based on frequently poor network-level data (such as packet loss rates). This simplistic rule is unreliable in determining the actual status of the equipment, resulting in relatively low reliability. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a method, device and equipment for monitoring the status of environmentally friendly public toilets, so as to improve the problem of relatively low reliability of equipment status monitoring in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solution:

[0005] A method for monitoring the status of environmentally friendly public toilet facilities includes:

[0006] The target environmentally friendly public toilet equipment is used to acquire multi-dimensional status time-series data, which includes network-level time-series data, equipment operation time-series data, and equipment environment time-series data. The target environmentally friendly public toilet equipment is used to interact with the backend equipment through the network.

[0007] Cross-dimensional state-time semantic mining is performed on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data. The multi-dimensional state-time features carry three dimensions of state-time semantic features corresponding to network-level time-time data, device runtime time-time data, and device environment time-time data.

[0008] Based on the multidimensional state time-series features, network state prediction is performed to obtain the target network state prediction result, wherein the target network state prediction result is used to reflect the probability of network failure of the target environmentally friendly public toilet equipment.

[0009] In a preferred embodiment of this application, the step of performing cross-dimensional state-time semantic mining on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data in the above-mentioned environmentally friendly public toilet equipment status monitoring method includes:

[0010] Semantic space mapping is performed on the network layer time-series data, the device runtime time-series data, and the device environment time-series data respectively to form network layer semantic features, device runtime semantic features, and device environment semantic features.

[0011] The network-level semantic features and the device operation semantic features are fused to form network device fusion features. Furthermore, based on the latent features represented by the device environment semantic features, the network device fusion features are subjected to salient feature screening to form multidimensional state temporal features.

[0012] In a preferred embodiment of this application, the steps of fusing the network-level semantic features and the device operation semantic features to form network device fusion features, and filtering the network device fusion features based on the latent features represented by the device environment semantic features to form multidimensional state temporal features, include:

[0013] The network-level semantic features and the device operation semantic features are concatenated to form network device fusion features;

[0014] The semantic features of the device environment are nonlinearly mapped to form latent features of the device environment;

[0015] Based on the hidden features of the device environment, significant feature screening is performed on the fusion features of the network devices to form multidimensional state temporal features.

[0016] In a preferred embodiment of this application, the step of performing nonlinear mapping on the semantic features of the equipment environment to form latent features of the equipment environment in the above-mentioned environmentally friendly public toilet equipment status monitoring method includes:

[0017] The semantic features of the device environment are processed by full connection to form fully connected features of the device environment;

[0018] Based on the first nonlinear function, the fully connected features of the device environment are processed to form nonlinear features of the device environment;

[0019] The nonlinear characteristics of the device environment are processed by full connectivity to form fully connected nonlinear characteristics;

[0020] Based on the second nonlinear function, the fully connected nonlinear features are processed to form nonlinear mapping features;

[0021] The hidden features of the device environment are determined based on the aforementioned nonlinear mapping features.

[0022] In a preferred embodiment of this application, in the aforementioned method for monitoring the status of public toilet equipment, the step of performing salient feature filtering on the network device fusion features based on the hidden environmental features of the equipment to form multidimensional status time-series features includes:

[0023] The hidden features of the device environment and the fused features of the network device are multiplied bitwise to form a multidimensional state temporal feature.

[0024] In a preferred embodiment of this application, in the above-mentioned method for monitoring the status of public toilet equipment, the step of performing semantic space mapping on the network-level time-series data, the equipment operation time-series data, and the equipment environment time-series data to form network-level semantic features, equipment operation semantic features, and equipment environment semantic features includes:

[0025] The network-level time-series data is processed by word embedding to form a word embedding feature sequence;

[0026] For each word embedding feature in the word embedding feature sequence, the current word embedding feature is multiplied by the first shared weight matrix to obtain the first intermediate feature. The output feature corresponding to the previous word embedding feature of the current word embedding feature is multiplied by the second shared weight matrix to obtain the second intermediate feature. The first intermediate feature, the second intermediate feature, and the shared bias parameter are added together to form the output feature of the current word embedding feature. The output feature corresponding to the previous word embedding feature of the first word embedding feature is a zero feature.

[0027] Based on the output features corresponding to the embedding features of the last word, the semantic features at the network level are determined;

[0028] Semantic space mapping is performed on the device runtime timing data and the device environment timing data respectively to form device runtime semantic features and device environment semantic features.

[0029] In a preferred embodiment of this application, the step of determining the network-level semantic features based on the output features corresponding to the embedding features of the last word in the above-mentioned environmentally friendly public toilet equipment status monitoring method includes:

[0030] Self-attention processing is applied to the output features corresponding to the embedding features of the last word to form semantic features at the network level.

[0031] In a preferred embodiment of this application, the above-mentioned method for monitoring the status of environmentally friendly public toilet equipment further includes:

[0032] Acquire training multidimensional state time-series data, wherein the multidimensional state time-series data includes training network layer time-series data, training device runtime time-series data, and training device environment time-series data;

[0033] Using the semantic mining unit included in the network state prediction model, cross-dimensional state temporal semantic mining is performed on the training multidimensional state temporal data to form training multidimensional state temporal features corresponding to the training multidimensional state temporal data. The training multidimensional state temporal features carry three dimensions of state temporal semantic features corresponding to the training network layer temporal data, the training device runtime temporal data, and the training device environment temporal data.

[0034] Using the state prediction unit included in the network state prediction model, network state prediction is performed based on the training multidimensional state temporal features to obtain the training network state prediction result.

[0035] Based on the error between the network state prediction results and the network state labels corresponding to the training multidimensional state time series data, the model parameters of the network state prediction model are updated to form an optimized network state prediction model.

[0036] This application also provides an environmentally friendly public toilet equipment status monitoring device, including:

[0037] The status data acquisition module is used to acquire multi-dimensional status time-series data of the target environmentally friendly public toilet equipment. The multi-dimensional status time-series data includes network layer time-series data, equipment operation time-series data, and equipment environment time-series data. The target environmentally friendly public toilet equipment is used to interact with the backend equipment through the network.

[0038] The semantic mining module is used to perform cross-dimensional state-time semantic mining on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data. The multi-dimensional state-time features carry three dimensions of state-time semantic features corresponding to network-level time-time data, device runtime time-time data, and device environment time-time data.

[0039] The state prediction module is used to predict the network state based on the multidimensional state time series features to obtain the target network state prediction result, wherein the target network state prediction result is used to reflect the probability of the target environmentally friendly public toilet equipment experiencing network failure.

[0040] Based on the above, this application also provides an electronic device, including:

[0041] Memory, used to store computer programs;

[0042] A processor connected to the memory is used to execute the computer program stored in the memory to implement the above-mentioned method for monitoring the status of environmentally friendly public toilet equipment.

[0043] The environmentally friendly public toilet equipment status monitoring method, device, and equipment provided in this application first acquire multi-dimensional state time-series data of the target environmentally friendly public toilet equipment; second, perform cross-dimensional state time-series semantic mining on the multi-dimensional state time-series data to form multi-dimensional state time-series features corresponding to the multi-dimensional state time-series data; then, perform network state prediction based on the multi-dimensional state time-series features to obtain the target network state prediction result. Based on the above, since cross-dimensional state time-series semantic mining is performed on the three dimensions of data—network-level time-series data, equipment operation time-series data, and equipment environment time-series data—the semantic features of the equipment operation and equipment environment dimensions can assist in representing the semantic information of the network level. In places like toilets, the humidity is relatively high, which can affect the lifespan of hardware equipment, and the moisture in the air can increase the propagation loss of wireless signals. Therefore, with the assistance of the semantic features of the equipment environment dimension, the semantic representation accuracy of the formed multi-dimensional state time-series features can be improved, thereby improving the reliability of the determined target network state prediction result, and thus improving the problem of relatively low reliability of equipment status monitoring in the prior art. Attached Figure Description

[0044] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings.

[0045] Figure 1 A structural block diagram of an electronic device provided in an embodiment of this application.

[0046] Figure 2 This is a flowchart illustrating the environmentally friendly public toilet equipment status monitoring method provided in this application embodiment.

[0047] Figure 3 This is a schematic diagram of semantic space mapping provided in an embodiment of this application.

[0048] Figure 4 A block diagram illustrating the status monitoring device for environmentally friendly public toilets provided in this application embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0050] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0051] like Figure 1 As shown in the figure, this application provides an electronic device. The electronic device may include a memory, a processor, and an environmentally friendly public toilet equipment status monitoring device.

[0052] In detail, the memory and the processor are electrically connected directly or indirectly to enable data transmission or interaction. For example, the memory and the processor can be electrically connected via one or more communication buses or signal lines. The environmentally friendly public toilet equipment status monitoring device includes at least one software functional module stored in the memory in the form of software or firmware. The processor is used to execute executable computer programs stored in the memory, such as the software functional modules and computer programs included in the environmentally friendly public toilet equipment status monitoring device, to implement the environmentally friendly public toilet equipment status monitoring method provided in this application embodiment.

[0053] Optionally, the memory may be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.

[0054] Furthermore, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), 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 gate or transistor logic devices, or discrete hardware components.

[0055] Understandable. Figure 1The structure shown is for illustrative purposes only; the electronic device may also include components that are more advanced than those shown. Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown may include, for example, a communication unit for exchanging information with other devices.

[0056] Combination Figure 2 This application also provides a method for monitoring the status of environmentally friendly public toilet equipment, applicable to the aforementioned electronic device. The method steps defined in the process of this environmentally friendly public toilet equipment status monitoring method can be implemented by the electronic device.

[0057] The following will be about Figure 2 The specific process shown will be explained in detail.

[0058] Step S110: Obtain multi-dimensional status time-series data of the target environmentally friendly public toilet equipment.

[0059] In this embodiment, the electronic device can acquire multi-dimensional state time-series data of the target eco-friendly public toilet device. This multi-dimensional state time-series data includes network-level time-series data, device runtime time-series data, and device environmental time-series data. The target eco-friendly public toilet device is used to interact with backend devices via a network. For example, the network-level time-series data may refer to at least one of packet loss rate, signal-to-noise ratio (SNR), and network latency at each of multiple time points. The device runtime time-series data may refer to the usage of at least one of processor, memory, and storage resources at each of multiple time points. The device environmental time-series data may refer to at least one of temperature and humidity at each of multiple time points.

[0060] Step S120: Perform cross-dimensional state-time semantic mining on the multidimensional state-time data to form multidimensional state-time features corresponding to the multidimensional state-time data.

[0061] In this embodiment, after acquiring the multidimensional state time-series data, the electronic device can perform cross-dimensional state time-series semantic mining on the multidimensional state time-series data to form multidimensional state time-series features corresponding to the multidimensional state time-series data. These multidimensional state time-series features carry three dimensions of state time-series semantic features corresponding to network-level time-series data, device runtime time-series data, and device environment time-series data. That is, within each dimension, not only are the semantic features of the data itself mined, but also the semantic relationships between data at different times are mined. Furthermore, the semantic features of the three dimensions are fused to form multidimensional state time-series features, thus achieving a balance between semantic richness and representational accuracy.

[0062] Step S130: Based on the multidimensional state temporal features, perform network state prediction to obtain the target network state prediction result.

[0063] In this embodiment, after forming the multidimensional state temporal features, the electronic device can perform network state prediction based on the multidimensional state temporal features to obtain a target network state prediction result. The target network state prediction result reflects the probability of a network failure occurring in the target environmentally friendly public toilet equipment. For example, the multidimensional state temporal features can be fully connected to obtain a 1*1 fully connected feature. Then, an identity mapping (e.g., f(x) = x) or a linear mapping (i.e., mapping to 0-1) can be performed on this fully connected feature to obtain the probability of a network failure, i.e., the target network state prediction result.

[0064] Based on the above, cross-dimensional state-time semantic mining is performed on data from three dimensions: network-level time-series data, device runtime time-series data, and device environment time-series data. This allows the semantic features of the device operation and device environment dimensions to assist in the representation of semantic information at the network level. In places like toilets, the humidity is relatively high, which can affect the lifespan of hardware devices, and the moisture in the air can increase the propagation loss of wireless signals. Therefore, by using the semantic features of the device environment dimension, the accuracy of the semantic representation of the multi-dimensional state-time features can be improved, thereby increasing the reliability of the predicted results of the determined target network state. This improves the problem of relatively low reliability of device state monitoring in existing technologies.

[0065] It should be noted that the specific method for cross-dimensional state time sequence semantic mining of the multidimensional state time sequence data is not limited and can be selected according to requirements.

[0066] For example, in an alternative implementation, step S120 described above may include:

[0067] First, semantic space mapping can be performed on the network-level time-series data, the device runtime time-series data, and the device environment time-series data respectively (in the process of semantic space mapping, it is necessary to capture the relationship between semantic features at different times) to form network-level semantic features, device runtime semantic features, and device environment semantic features; second, the network-level semantic features, device runtime semantic features, and device environment semantic features can be concatenated and self-attention processed to obtain the multidimensional state time-series features corresponding to the multidimensional state time-series data.

[0068] For example, in another alternative implementation, in order to achieve reliable fusion of multidimensional semantic features and thereby further improve the semantic representation accuracy of the formed multidimensional state temporal features, the above step S120 may further include steps S121 and S122, the specific contents of which are as follows.

[0069] Step S121: Semantic space mapping is performed on the network layer time-series data, the device runtime time-series data, and the device environment time-series data respectively to form network layer semantic features, device runtime semantic features, and device environment semantic features.

[0070] In this embodiment, the network-level time-series data, the device runtime time-series data, and the device environment time-series data can be semantically mapped to form network-level semantic features, device runtime semantic features, and device environment semantic features, respectively. That is, the network-level time-series data can be mapped to a semantic space for representation, thereby forming corresponding network-level semantic features. Similarly, the device runtime time-series data can be mapped to a semantic space for representation, thereby forming corresponding device runtime semantic features. And the device environment time-series data can be mapped to a semantic space for representation, thereby forming corresponding device environment semantic features. It should be noted that the representation of each semantic feature can be a vector (or matrix), such as:

[0071] [0.54, 0.76, 0.19, 0.96, 0.40, 0.13, 0.87, 0.60, 0.54, 0.25, 0.86, 0.35, 0.06, 0.11, 0.80, 0.26, 0.94, 0.67, 0.48, ..., 0.52].

[0072] Step S122: The network-level semantic features and the device operation semantic features are fused to form network device fusion features. The network device fusion features are then subjected to salient feature screening based on the latent features represented by the device environment semantic features to form multidimensional state temporal features.

[0073] In this embodiment, the network-level semantic features and the device operation semantic features can be fused to form network device fused features. Furthermore, based on the latent features represented by the device environment semantic features, salient feature filtering is performed on the network device fused features to form multidimensional state temporal features. It should be noted that since the semantic features of the device operation dimension and the semantic features of the network layer are more closely related, they can be fused first, and then fused again using the semantic features of the device environment dimension, which have a certain potential relationship.

[0074] It is understood that in step S121 above, the specific methods for semantic space mapping of the network layer timing data, the device runtime timing data, and the device environment timing data are not limited. For example, in an alternative implementation, in order to fully capture the semantic information about timing between the data in addition to mining the semantic information of the data itself, step S121 above may further include steps S121a, S121b, S121c, and S121d, the specific contents of each step are as follows.

[0075] Step S121a: Perform word embedding processing on the network-level time-series data to form a word embedding feature sequence.

[0076] In this embodiment, word embedding processing (such as using a word embedding model like Word2Vec) can be performed on the network-level time-series data to form a word embedding feature sequence. It should be noted that when each of the multiple time points included in the network-level time-series data corresponds to multiple data types, such as packet loss rate, signal-to-noise ratio, and network latency, word embedding processing can be performed on each type of data to obtain corresponding local word embedding features. Then, the mean of the local word embedding features corresponding to each data type can be calculated to obtain the corresponding word embedding feature sequence. That is, in the word embedding feature sequence, each word embedding feature corresponds to a time point, and the word embedding features can be sorted according to the chronological relationship between the corresponding time points, such as from early to late.

[0077] Step S121b: For each word embedding feature in the word embedding feature sequence, multiply the current word embedding feature by the first shared weight matrix to obtain the first intermediate feature; multiply the output feature corresponding to the previous word embedding feature of the word embedding feature by the second shared weight matrix to obtain the second intermediate feature; and add the first intermediate feature, the second intermediate feature and the shared bias parameter to form the output feature of the current word embedding feature.

[0078] In this embodiment, after obtaining the word embedding feature sequence, for each word embedding feature in the sequence, the current word embedding feature is multiplied by a first shared weight matrix to obtain a first intermediate feature; the output feature corresponding to the previous word embedding feature of the current word embedding feature is multiplied by a second shared weight matrix to obtain a second intermediate feature; and the first intermediate feature, the second intermediate feature, and the shared bias parameter are added to form the output feature of the current word embedding feature. The output feature corresponding to the previous word embedding feature of the first word embedding feature is a zero-valued feature, such as [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ..., 0]. That is, multiple word embedding features can be fused sequentially through multiple time steps, for example, by combining... Figure 3 In the first time step, the first word embedding feature and the all-zero feature can be fused. In the second time step, the second word embedding feature and the output feature of the first time step can be fused. In the third time step, the third word embedding feature and the output feature of the second time step can be fused. Feature fusion can be achieved using the first shared weight matrix, the second shared weight matrix, and the shared bias parameter. "Shared" means that the weight matrix and bias parameter are the same at each time step. Furthermore, to introduce non-linearity and capture non-linear features in semantic features, after adding the shared bias parameter, a non-linear activation process can be applied to the result to form the output feature of the current word embedding feature. The specific fusion formula can be:

[0079] H t =tanh(W h *H t-1 +W x *x t +b);

[0080] Among them, H t Let W be the output feature of the current word embedding, tanh be the function for non-linear activation, and W be the output feature of the current word embedding. h H is the second shared weight matrix. t-1 W is the output feature corresponding to the embedding feature of the previous word. x Let x be the first shared weight matrix. t is the current word embedding feature, and b is the shared bias parameter.

[0081] Step S121c: Based on the output features corresponding to the embedding features of the last word, determine the semantic features at the network level.

[0082] In this embodiment, after forming the output feature corresponding to the last word embedding feature, the network-level semantic features can be determined based on the output feature corresponding to the last word embedding feature. For example, in an alternative implementation, the output feature corresponding to the last word embedding feature can be used as the network-level semantic features. Another alternative implementation involves performing self-attention processing on the output feature corresponding to the last word embedding feature (i.e., transforming the output feature linearly to generate three vectors: Q (query vector), K (key vector), and V (value vector); calculating the dot product of Q and K, then performing scaling and softmax operations to obtain the attention weight matrix A; and weighting and summing the attention weight matrix A and the value vector V to obtain the final output vector O, i.e., the network-level semantic features), thus forming the network-level semantic features.

[0083] Step S121d: Semantic space mapping is performed on the device runtime timing data and the device environment timing data respectively to form device runtime semantic features and device environment semantic features.

[0084] In this embodiment of the application, the device runtime timing data and the device environment timing data can be semantically mapped to form device runtime semantic features and device environment semantic features respectively. The specific processing method can refer to the process of processing the network layer timing data.

[0085] It is understood that the specific method of forming multidimensional state temporal features in step S122 above is not limited. For example, in an alternative implementation, in order to extract important semantic features by fusing the semantic features of the three dimensions, step S122 above may further include steps S122a, S122b and S122c, the specific contents of each step are as follows.

[0086] Step S122a: The network-level semantic features and the device operation semantic features are concatenated to form network device fusion features.

[0087] In this embodiment, the network-level semantic features and the device operational semantic features can be concatenated to form a network device fusion feature. The size of both the network-level semantic features and the device operational semantic features can be 1*n, so the size of the network device fusion feature can be 1*m, where m=2n.

[0088] Step S122b: Perform nonlinear mapping on the semantic features of the device environment to form latent features of the device environment.

[0089] In this embodiment of the application, the semantic features of the device environment can be nonlinearly mapped to form latent features of the device environment, that is, to capture the nonlinear information in the semantic features of the device environment.

[0090] Step S122c: Based on the hidden features of the device environment, perform significant feature filtering on the network device fusion features to form multi-dimensional state temporal features.

[0091] In this embodiment, after forming the device environment latent features and the network device fusion features, significant feature filtering can be performed on the network device fusion features based on the device environment latent features to form multidimensional state-time features, thus achieving the filtering of important features. For example, to achieve significant feature filtering in the network device fusion features, the device environment features and the network device fusion features can be multiplied bitwise to form multidimensional state-time features, thereby allowing attention to important features in the network device fusion features. Furthermore, it should be noted that since the network device fusion feature is actually formed by concatenating the network-level semantic features and the device operation semantic features, after salient feature filtering based on the device environment latent features, i.e. after bitwise multiplication, the important features of the network-level semantic features and the device operation semantic features are actually captured. Therefore, the full fusion between the two is not achieved. Thus, in an alternative implementation, self-attention processing can be applied to the result of the bitwise multiplication operation to cross-fuse the important features of the two dimensions, thereby capturing the associated semantic information and further improving the accuracy of the multidimensional state temporal features.

[0092] It is understood that the specific method of performing nonlinear mapping on the device environment semantic features in step S122b above is not limited. For example, in an alternative implementation, in order to effectively capture the nonlinear high-order semantic features in the device environment semantic features, step S122b above may further include the following:

[0093] First, the device environment semantic features can be processed by full connection to form device environment full connection features. For example, the device environment semantic features can be multiplied by the first full connection weight matrix, and then the result of the multiplication can be added to the first full connection bias parameter to obtain the device environment full connection features.

[0094] Secondly, the fully connected features of the device environment can be processed based on the first nonlinear function to form nonlinear features of the device environment. For example, the first nonlinear function can be a modified linear unit (ReLU).

[0095] Then, the nonlinear features of the device environment can be processed by full connection to form fully connected nonlinear features. For example, the nonlinear features of the device environment can be multiplied by the second fully connected weight matrix, and then the result of the multiplication can be added to the second fully connected bias parameter to obtain the fully connected nonlinear features.

[0096] Furthermore, the fully connected nonlinear features can be processed based on the second nonlinear function to form nonlinear mapping features. For example, the first nonlinear function can be sigmoid, which is used to map each feature parameter to the interval (0, 1).

[0097] Finally, the device environment latent features can be determined based on the nonlinear mapping features. For example, the nonlinear mapping features can be used as device environment latent features; or, the nonlinear mapping features can be further processed, for example, multiplied by a scaling factor to obtain the device environment latent features, which can be a value such as 2 or 3.

[0098] Further explanation is needed regarding steps S120 and S130 above. To ensure the reliable implementation of semantic mining and state prediction, the corresponding steps can be executed using a trained neural network model. The training process of the neural network model is as follows:

[0099] First, training multidimensional state time series data can be obtained, wherein the multidimensional state time series data includes training network layer time series data, training device runtime time series data, and training device environment time series data. Please refer to the relevant explanations and descriptions of the multidimensional state time series data above.

[0100] Secondly, the semantic mining unit included in the network state prediction model can be used to perform cross-dimensional state temporal semantic mining on the training multidimensional state temporal data to form training multidimensional state temporal features corresponding to the training multidimensional state temporal data. The training multidimensional state temporal features carry three dimensions of state temporal semantic features corresponding to the training network layer temporal data, the training device runtime temporal data, and the training device environment temporal data. Refer to the previous explanation of step S120. The first shared weight matrix, the second shared weight matrix, the shared bias parameter, and the parameters corresponding to attention processing can be randomly generated initially.

[0101] Then, the state prediction unit included in the network state prediction model can be used to predict the network state based on the training multidimensional state temporal features to obtain the training network state prediction result. Refer to the relevant explanation of step S130 above.

[0102] Finally, the model parameters of the network state prediction model can be updated based on the error (such as mean square error) between the training network state prediction results and the network state labels corresponding to the training multidimensional state time series data, to form an optimized network state prediction model. For example, the model parameters can be adjusted in the direction of reducing the error until the error converges.

[0103] Combination Figure 4 This application also provides an environmentally friendly public toilet equipment status monitoring device applicable to the aforementioned electronic devices. The environmentally friendly public toilet equipment status monitoring device may include a status data acquisition module, a semantic mining module, and a status prediction module.

[0104] Specifically, the status data acquisition module can be used to acquire multi-dimensional status time-series data of the target environmentally friendly public toilet equipment. This multi-dimensional status time-series data includes network-level time-series data, equipment runtime time-series data, and equipment environmental time-series data. The target environmentally friendly public toilet equipment is used to interact with backend equipment via the network. In this embodiment, the status data acquisition module can be used to execute... Figure 2 The relevant content regarding the status data acquisition module in step S110 shown can be found in the previous description of step S110.

[0105] In detail, the semantic mining module can be used to perform cross-dimensional state-time semantic mining on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data. These multi-dimensional state-time features carry three dimensions of state-time semantic features corresponding to network-level time-time data, device runtime time-time data, and device environment time-time data. In this embodiment, the semantic mining module can be used to execute... Figure 2 The relevant content regarding the semantic mining module in step S120 shown can be found in the previous description of step S120.

[0106] In detail, the state prediction module can be used to perform network state prediction based on the multi-dimensional state time-series features to obtain a target network state prediction result, wherein the target network state prediction result is used to reflect the probability of network failure of the target environmentally friendly public toilet equipment. In the embodiments of this application, the state prediction module can be used to perform... Figure 2 The relevant content regarding the state prediction module in step S130 shown can be found in the previous description of step S130.

[0107] In this embodiment of the application, corresponding to the above-described method for monitoring the status of environmentally friendly public toilet equipment applied to the electronic device, a computer-readable storage medium is also provided, which stores a computer program that executes the various steps of the method for monitoring the status of environmentally friendly public toilet equipment when the computer program is run.

[0108] The steps executed by the aforementioned computer program during runtime will not be described in detail here, but can be found in the explanation of the environmentally friendly public toilet equipment status monitoring method described above.

[0109] In summary, the environmentally friendly public toilet equipment status monitoring method, device, and equipment provided in this application first acquire multi-dimensional state time-series data of the target environmentally friendly public toilet equipment; second, perform cross-dimensional state time-series semantic mining on the multi-dimensional state time-series data to form multi-dimensional state time-series features corresponding to the multi-dimensional state time-series data; and then, perform network state prediction based on the multi-dimensional state time-series features to obtain the target network state prediction result. Based on the above, because cross-dimensional state time-series semantic mining is performed on the three dimensions of data—network-level time-series data, equipment operation time-series data, and equipment environment time-series data—the semantic features of the equipment operation and equipment environment dimensions can assist in representing the semantic information at the network level. In places like toilets, humidity is relatively high, which can affect the lifespan of hardware equipment, and moisture in the air increases the propagation loss of wireless signals. Therefore, with the assistance of semantic features from the equipment environment dimension, the semantic representation accuracy of the formed multi-dimensional state time-series features can be improved, thereby increasing the reliability of the determined target network state prediction result, and thus improving the problem of relatively low reliability of equipment status monitoring in existing technologies.

[0110] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus and method embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0111] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0112] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, electronic device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further restrictions, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0113] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for monitoring the status of environmentally friendly public toilet equipment, characterized in that, include: The target environmentally friendly public toilet equipment is used to acquire multi-dimensional status time-series data, which includes network-level time-series data, equipment operation time-series data, and equipment environment time-series data. The target environmentally friendly public toilet equipment is used to interact with the backend equipment through the network. Cross-dimensional state-time semantic mining is performed on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data. The multi-dimensional state-time features carry three dimensions of state-time semantic features corresponding to network-level time-time data, device runtime time-time data, and device environment time-time data. Based on the multidimensional state time-series features, network state prediction is performed to obtain the target network state prediction result, wherein the target network state prediction result is used to reflect the probability of network failure of the target environmentally friendly public toilet equipment.

2. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 1, characterized in that, The step of performing cross-dimensional state-time semantic mining on the multidimensional state-time data to form multidimensional state-time features corresponding to the multidimensional state-time data includes: Semantic space mapping is performed on the network layer time-series data, the device runtime time-series data, and the device environment time-series data respectively to form network layer semantic features, device runtime semantic features, and device environment semantic features. The network-level semantic features and the device operation semantic features are fused to form network device fusion features. Furthermore, based on the latent features represented by the device environment semantic features, the network device fusion features are subjected to salient feature screening to form multidimensional state temporal features.

3. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 2, characterized in that, The steps of fusing the network-level semantic features and the device operational semantic features to form network device fused features, and filtering the network device fused features based on the latent features represented by the device environment semantic features to form multidimensional state temporal features, include: The network-level semantic features and the device operation semantic features are concatenated to form network device fusion features; The semantic features of the device environment are nonlinearly mapped to form latent features of the device environment; Based on the hidden features of the device environment, significant feature screening is performed on the fusion features of the network devices to form multidimensional state temporal features.

4. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 3, characterized in that, The step of performing nonlinear mapping on the semantic features of the device environment to form latent features of the device environment includes: The semantic features of the device environment are processed by full connection to form fully connected features of the device environment; Based on the first nonlinear function, the fully connected features of the device environment are processed to form nonlinear features of the device environment, wherein the first nonlinear function is a modified linear unit; The nonlinear characteristics of the device environment are processed by full connectivity to form fully connected nonlinear characteristics; Based on the second nonlinear function, the fully connected nonlinear features are processed to form nonlinear mapping features, wherein the second nonlinear function is sigmoid; The hidden features of the device environment are determined based on the aforementioned nonlinear mapping features.

5. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 3, characterized in that, The step of performing salient feature filtering on the network device fusion features based on the device environment latent features to form multidimensional state temporal features includes: The hidden features of the device environment and the fused features of the network device are multiplied bitwise to form a multidimensional state temporal feature.

6. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 2, characterized in that, The step of performing semantic space mapping on the network-level time-series data, the device runtime time-series data, and the device environment time-series data to form network-level semantic features, device runtime semantic features, and device environment semantic features includes: The network-level time-series data is processed by word embedding to form a word embedding feature sequence; For each word embedding feature in the word embedding feature sequence, the current word embedding feature is multiplied by the first shared weight matrix to obtain the first intermediate feature. The output feature corresponding to the previous word embedding feature of the current word embedding feature is multiplied by the second shared weight matrix to obtain the second intermediate feature. The first intermediate feature, the second intermediate feature, and the shared bias parameter are added together to form the output feature of the current word embedding feature. The output feature corresponding to the previous word embedding feature of the first word embedding feature is a zero feature. Based on the output features corresponding to the embedding features of the last word, the semantic features at the network level are determined; Semantic space mapping is performed on the device runtime timing data and the device environment timing data respectively to form device runtime semantic features and device environment semantic features.

7. The method for monitoring the status of environmentally friendly public toilet equipment according to claim 6, characterized in that, The step of determining the network-level semantic features based on the output features corresponding to the last word embedding features includes: Self-attention processing is applied to the output features corresponding to the embedding features of the last word to form semantic features at the network level.

8. The method for monitoring the status of environmentally friendly public toilet equipment according to any one of claims 1-7, characterized in that, The method for monitoring the status of environmentally friendly public toilet equipment also includes: Acquire training multidimensional state time-series data, wherein the multidimensional state time-series data includes training network layer time-series data, training device runtime time-series data, and training device environment time-series data; Using the semantic mining unit included in the network state prediction model, cross-dimensional state temporal semantic mining is performed on the training multidimensional state temporal data to form training multidimensional state temporal features corresponding to the training multidimensional state temporal data. The training multidimensional state temporal features carry three dimensions of state temporal semantic features corresponding to the training network layer temporal data, the training device runtime temporal data, and the training device environment temporal data. Using the state prediction unit included in the network state prediction model, network state prediction is performed based on the training multidimensional state temporal features to obtain the training network state prediction result. Based on the error between the network state prediction results and the network state labels corresponding to the training multidimensional state time series data, the model parameters of the network state prediction model are updated to form an optimized network state prediction model.

9. A status monitoring device for environmentally friendly public toilet equipment, characterized in that, include: The status data acquisition module is used to acquire multi-dimensional status time-series data of the target environmentally friendly public toilet equipment. The multi-dimensional status time-series data includes network layer time-series data, equipment operation time-series data, and equipment environment time-series data. The target environmentally friendly public toilet equipment is used to interact with the backend equipment through the network. The semantic mining module is used to perform cross-dimensional state-time semantic mining on the multi-dimensional state-time data to form multi-dimensional state-time features corresponding to the multi-dimensional state-time data. The multi-dimensional state-time features carry three dimensions of state-time semantic features corresponding to network-level time-time data, device runtime time-time data, and device environment time-time data. The state prediction module is used to predict the network state based on the multidimensional state time series features to obtain the target network state prediction result, wherein the target network state prediction result is used to reflect the probability of the target environmentally friendly public toilet equipment experiencing network failure.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor connected to the memory is used to execute the computer program stored in the memory to implement the environmentally friendly public toilet equipment status monitoring method according to any one of claims 1-8.

Images