Urban Environmental Quality Assessment by BP Neural Network Model

The urban environmental quality situation is becoming much serious with the development of economy and society, especially with the urbanization. Urban environmental quality assessment result is key background material to make urban environmental protection plan. A neural network model is useful in such cases because of it self-learning capability. This study presented a novel approach that applied a BP (back propagation) neural network model as a tool to evaluate the decision-making process in city environmental quality assessment. The proposed model was designed to use monitoring parameters to construct the neural network model. The components of input layer consisted of six parameters including the waste water, waste gas, waste solid, sulfur dioxide, TSP and noise. Assessment grades, consisting of three grades, are defined as the outputs in the modeling analysis. To ease the computational effort, a single hidden layer with different count of nodes was selected. The case study of Zhaotong city environmental monitoring data reported herein show that reliable and beneficial assessment method was attainable. Back propagation network algorithm was favored for such applications. Research findings indicated that a well- trained neural network model was helpful in the assessment of urban environmental quality practice to achieve high precision.