| CPC G06Q 10/06315 (2013.01) [G16Y 10/35 (2020.01)] | 12 Claims |
|
1. A method for maintenance management of a smart gas call center, implemented based on an Internet of Things (IoT) system for maintenance management of a smart gas call center, wherein the IoT system includes a smart gas user platform, a smart gas service platform, a smart gas management platform, a smart gas sensor network platform, and a smart gas object platform which interact in sequence, wherein
the smart gas management platform includes a smart customer service management sub-platform, a smart operation management sub-platform, and a smart gas data center, the smart customer service management sub-platform bidirectionally interacts with the smart gas data center, the smart operation management sub-platform bidirectionally interacts with the smart gas data center, and the smart customer service management sub-platform and the smart operation management sub-platform obtain data from the smart gas data center and feedback corresponding operation information;
the smart gas user platform includes a gas user sub-platform, a government user sub-platform, and a supervision user sub-platform, the gas user sub-platform corresponds to a gas user, the government user sub-platform corresponds to a government user, and the supervision user sub-platform corresponds to a supervision user;
the smart gas service platform includes a smart gas usage service sub-platform, a smart operation service sub-platform, and a smart supervision service sub-platform, the smart gas usage service sub-platform corresponds to the gas user sub-platform, the smart operation service sub-platform corresponds to the government user sub-platform, and the smart supervision service sub-platform corresponds to the supervision user sub-platform;
the smart gas object platform includes a gas indoor device object sub-platform and a gas pipeline network device object sub-platform, the gas indoor device object sub-platform includes a metering device of the gas user, the gas pipeline network device object sub-platform includes at least one of a pressure regulation device, a gas gate station compressor, a gas flow meter, a valve control device, a thermometer, and a barometer; and
the method is performed by a processor of the smart gas management platform, the method comprising:
obtaining a call consultation data information of the gas user through the smart gas service platform based on the smart gas user platform and the smart gas object platform;
generating a second data information from the call consultation data information, the second data information including relevant data information generated by a fault location of an agent to be evaluated within a target time period;
generating one or more location feature vectors of the agent to be evaluated based on the second data information;
calculating a first location evaluation value and a location accuracy for each location feature vector in each dimension based on the one or more location feature vectors; and
performing a weighted summation on a plurality of first location evaluation values obtained by calculating the one or more location feature vectors in the each dimension to generate a multi-dimensional location evaluation value, a weight of the weighted summation being related to a corresponding location accuracy, a dimension of a maintenance feature vector being determined based on a location complexity corresponding to the fault location, and the location complexity corresponding to the fault location being generated based on a model output ambiguity corresponding to the each location feature vector and a gas fault distribution, wherein each of a plurality of dimensions corresponds to the location complexity, each of the one or more location feature vectors corresponds to the fault location, and the agent to be evaluated locates a gas fault type based on the call consultation data information, wherein a manner for generating the location accuracy includes:
determining user-side gas feature data, gas composition features, gas entry features, and gas upstream transportation features based on the second data information, the gas composition features referring to a gas composition, the gas entry features referring to pressure, flow velocity, and a temperature of gas at a plurality of points in a gas entry section, the gas upstream transportation features referring to pressure, flow velocity, and a temperature of gas at a plurality of points in an upstream gas transportation section, wherein the gas composition features, the gas entry features, and the gas upstream transportation features are obtained by the gas flow meter, the thermometer, and the barometer of the smart gas object platform;
inputting the user-side gas feature data, the gas composition features, the gas entry features, and the gas upstream transportation features into a gas fault prediction model, and processing the user-side gas feature data, the gas composition features, the gas entry features, and the gas upstream transportation features using the gas fault prediction model to output a predicted gas fault type, the gas fault prediction model being a machine learning model, wherein
the gas fault prediction model is trained based on a plurality of labeled training samples, the training samples include the user-side gas feature data, the gas composition features, the gas entry features, and the gas upstream transportation features of historical gas users, training labels include known gas fault types of the historical gas users, wherein a training process of the gas fault prediction model includes:
inputting the plurality of labeled training samples into an initial gas fault prediction model;
constructing a loss function through the training labels and output of the initial gas fault prediction model;
iteratively updating parameters of the initial gas fault prediction model through gradient descent based on the loss function to obtain a trained gas fault prediction model; and
completing model training and obtaining the trained gas fault prediction model until the loss function of the initial gas fault prediction model meeting a preset condition, wherein the preset iteration condition includes convergence of the loss function and a number of iterations reaching a threshold; and
generating the location accuracy based on the predicted gas fault type and an actual gas fault type;
generating a multi-dimensional maintenance evaluation value of a maintainer to be evaluated based on the call consultation data information, wherein the maintainer to be evaluated performs maintenance based on the gas fault type;
adjusting a work order processing scope of the maintainer to be evaluated and the agent to be evaluated based on the multi-dimensional maintenance evaluation value and the multi-dimensional location evaluation value; and
managing a pipeline network project by the smart operation management sub-platform based on the adjusted work order processing scope of the maintainer to be evaluated and the agent to be evaluated.
|