An error evaluation method of temperature on-site measurement based on a MGF-SR-GA-BP model

Because currently available existing temperature sensors cannot be disassembled for measurement, we proposed an on-site testing method based on a MGF-SR-GA-BP model that integrates a Mean Generation Function, Stepwise Regression, a Genetic Algorithm, and Back Propagation. First, error data obtained through comparison with a standard device were periodically extended using a Mean Generation Function, and Stepwise Regression was applied to establish a MGF-SR model. Second, a Back Propagation neural network was used to establish a MGF-BP model, and a Genetic Algorithm was used to perform optimization and create an MGF-GA-BP model. Finally, a linear online evaluation model, namely the MGF-SR-GA-BP model, was established using the aforementioned models. This model uses the sum of squared prediction errors as the basis for its objective function. In experiments, compared with the MGF-SR and MGF-GA-BP models, the hybrid MGF-SR-GA-BP model achieved a lower average relative error (0.27 and 0.66 vs. 0.21, respectively). Unlike traditional temperature sensor testing models, the developed model can be used for on-site testing of temperature sensors because it can effectively handle unstable temperature environments. This model is effective for evaluating errors when equipment cannot be feasibly disassembled to enable measurement against a stable standardized testing device.