Methods for determining smart gas inspection plans and internet of things systems thereof

1. A method for determining a smart gas inspection plan, wherein the method is implemented by at least one processor of an internet of things system, wherein the internet of things system further comprises a smart gas pipeline network safety management platform, a smart gas user platform, a smart gas service platform, a smart gas sensor network platform, and a smart gas object platform, wherein the smart gas pipeline network safety management platform includes a smart gas pipeline network inspection management sub-platform and a smart gas data center, the smart gas object platform includes a smart gas pipeline network inspection project object sub-platform, and the method comprises:

obtaining an area to be inspected through the smart gas data center;

determining one or more downstream users based on the area to be inspected, and obtaining historical gas consumption data of each of the one or more downstream users through the smart gas sensor network platform;

independently predicting future gas consumption data at a plurality of future times of the each of the one or more downstream users based on the historical gas consumption data of the each of the one or more downstream users through a prediction model, the prediction model being a machine learning model that is a neural network, wherein the prediction model is trained by a training process based on a plurality of first training samples with first labels, the training process including:

obtaining the first training samples based on historical data, wherein the first training samples includes historical gas consumption velocities at a plurality of historical times;

inputting the plurality of the first training samples with the first labels into an initial prediction model, wherein the first labels include actual gas consumption velocities corresponding to the historical gas consumption velocities at the plurality of historical times;

constructing a loss function based on the first labels and results of the initial prediction model;

updating parameters of the initial prediction model iteratively by gradient descent based on the loss function; and

obtaining the prediction model when a preset training condition is satisfied, wherein the preset training condition includes that the loss function converges or a number of iterations reaches a threshold;

independently determining a plurality of peak-valley sub-features of gas consumption of the each of the one or more downstream users at the future time based on the future gas consumption data at the plurality of future times of the each of the one or more downstream users;

determining the peak-valley features of gas consumption based on the plurality of peak-valley sub-features of gas consumption, wherein a process for determining the peak-valley features of gas consumption based on the plurality of peak-valley sub-features of gas consumption includes a weighted calculation manner, a weight of the weighted calculation manner is determined based on an inspection influence degree, the inspection influence degree refers to an influence degree of inspection on gas consumption of the gas user, and a process for determining the inspection influence degree includes:

constructing a gas transmission and distribution network graph, wherein positions with specific functions in the gas transmission and distribution network graph are abstracted as the nodes, the nodes include a gas source node, a pressure regulation node, a pipeline intersection node, and a user node, and gas transmission and distribution pipelines are abstracted as edges;

determining accessibility based on the gas transmission and distribution network graph, wherein the accessibility is used to represent whether the gas source node is capable of normally supplying gas to the user node;

determining the inspection influence degree based on the accessibility;

determining, based on the peak-valley features of gas consumption at the future time, an inspection time of the area to be inspected; and

generating inspection reminder instructions based on the inspection time through an inspection time warning module of the smart gas pipeline network inspection management sub-platform, wherein the inspection reminder instructions are sent to a corresponding smart gas pipeline network inspection project object sub-platform by the smart gas data center through a smart gas pipeline network inspection project sensor network sub-platform of the smart gas sensor network platform.

2. The method according to claim 1 , wherein the determining, based on the peak-valley features of gas consumption at the future time, an inspection time of the area to be inspected includes:

determining a plurality of optional time points based on the peak-valley features of gas consumption;

determining a target inspection time point based on the plurality of optional time points; and

determining the target inspection time point as a next inspection time.

3. The method according to claim 2 , wherein the determining a plurality of optional time points based on the peak-valley features of gas consumption includes:

determining a plurality of feature time points satisfying a first preset condition based on the peak-valley features of gas consumption; and

selecting a plurality of time points satisfying a second preset condition around each of the plurality of feature time points as the plurality of optional time points, wherein a count of the plurality of optional time points is related to a valley value of a valley point corresponding to the plurality of feature time points.

4. The method according to claim 2 , wherein the determining a target inspection time point based on the plurality of optional time points includes:

obtaining gas consumption data of the each of the one or more downstream users before the plurality of optional time points;

predicting future gas consumption features by processing the gas consumption data of the each of the one or more downstream users before the plurality of optional time points through the prediction model, wherein the future gas consumption features are gas consumption features after inspection based on each of the plurality of optional time points; and

determining the target inspection time point based on the future gas consumption features.

5. An internet of things system for determining a smart gas inspection plan, wherein the internet of things system comprises a smart gas user platform, a smart gas service platform, a smart gas pipeline network safety management platform, a smart gas sensor network platform, and a smart gas object platform, wherein the smart gas pipeline network safety management platform includes a smart gas pipeline network inspection management sub-platform and a smart gas data center, the smart gas object platform includes a smart gas pipeline network inspection project object sub-platform, wherein the Internet of Things system further comprises:

a non-transitory computer-readable storage medium storing computer instructions; and

at least one processor in communication with the non-transitory computer-readable storage medium, when executing the computer instructions, the at least one processor is directed to cause the Internet of Things system to perform operations including:

obtaining an area to be inspected through the smart gas data center;

determining one or more downstream users based on the area to be inspected, and obtaining historical gas consumption data of each of the one or more downstream users through the smart gas sensor network platform;

independently predicting future gas consumption data at the plurality of future times of the each of the one or more downstream users based on the historical gas consumption data of the each of the one or more downstream users through a prediction model, the prediction model being a machine learning model that is a neural network, wherein the prediction model is trained by a training process based on a plurality of first training samples with first labels, the training process including:

obtaining the first training samples based on historical data, wherein the first training samples includes historical gas consumption velocities at a plurality of historical times;

inputting the plurality of the first training samples with the first labels into an initial prediction model, wherein the first labels include actual gas consumption velocities corresponding to the historical gas consumption velocities at the plurality of historical times;

constructing a loss function based on the first labels and results of the initial prediction model;

updating parameters of the initial prediction model iteratively by gradient descent based on the loss function;

obtaining the prediction model when a preset training condition is satisfied, wherein the preset training condition includes that the loss function converges or a number of iterations reaches a threshold;

independently determining a plurality of peak-valley sub-features of gas consumption of the each of the one or more downstream users at the future time based on the future gas consumption data at the plurality of future times of the each of the one or more downstream users;

determining the peak-valley features of gas consumption based on the plurality of peak-valley sub-features of gas consumption, wherein a process for determining the peak-valley features of gas consumption based on the plurality of peak-valley sub-features of gas consumption includes a weighted calculation manner, a weight of the weighted calculation manner is determined based on an inspection influence degree, the inspection influence degree refers to an influence degree of inspection on gas consumption of the gas user, and a process for determining the inspection influence degree includes:

constructing a gas transmission and distribution network graph, wherein positions with specific functions in the gas transmission and distribution network graph are abstracted as the nodes, the nodes include a gas source node, a pressure regulation node, a pipeline intersection node, and a user node, and gas transmission and distribution pipelines are abstracted as edges;

determining accessibility based on the gas transmission and distribution network graph, wherein the accessibility is used to represent whether the gas source node is capable of normally supplying gas to the user node;

determining the inspection influence degree based on the accessibility;

determining, based on the peak-valley features of gas consumption at the future time, an inspection time of the area to be inspected; and

generating inspection reminder instructions based on the inspection time through an inspection time warning module of the smart gas pipeline network inspection management sub-platform, wherein the inspection reminder instructions are sent to a corresponding smart gas pipeline network inspection project object sub-platform by the smart gas data center through a smart gas pipeline network inspection project sensor network sub-platform of the smart gas sensor network platform.

6. The internet of things system according to claim 5 , wherein at least one processor is directed to cause the internet of things system to perform operations including:

determining a plurality of optional time points based on the peak-valley features of gas consumption;

determining a target inspection time point based on the plurality of optional time points; and

determining the target inspection time point as a next inspection time.

7. The internet of things system according to claim 6 , wherein the at least one processor is directed to cause the internet of things system to perform operations including:

determining a plurality of feature time points satisfying a first preset condition based on the peak-valley features of gas consumption; and

selecting a plurality of time points satisfying a second preset condition around each of the plurality of feature time points as the plurality of optional time points; wherein a count of the plurality optional time points is related to a valley value of a valley point corresponding to the plurality of feature time points.

8. The internet of things system according to claim 6 , wherein the at least one processor is directed to cause the internet of things system to perform operations including:

obtaining the gas consumption data of the each of the one or more downstream users before the plurality of optional time points;

predicting future gas consumption features by processing the gas consumption data of the each of the one or more downstream users before the plurality of optional time points through the prediction model, wherein the future gas consumption features are gas consumption features after inspection based on each of the plurality of optional time points; and

determining the target inspection time point based on the future gas consumption features.