Memory, biochemical oxygen demand soft measurement method, system and device
A biochemical oxygen demand and soft-sensing technology, applied in neural learning methods, design optimization/simulation, biological neural network models, etc., can solve the problems of prone to deviation, unstable measurement result accuracy, and inability of prediction models to adapt to application scenarios, etc. problem, to achieve the effect of improving accuracy and stability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0075] In order to make the measurement model of biochemical oxygen demand soft measurement in practical application, the measurement results generated by it are more accurate and stable, such as figure 1 As shown, a biochemical oxygen demand soft-sensing method is provided in the embodiment of the present invention. When establishing the biochemical oxygen demand BOD soft-sensing model based on the RBF neural network, the initial neural network of the hidden layer is determined during each PCA calculation. Number of elements, including steps:
[0076] S11. Obtain the input layer variables corresponding to the BOD soft sensor model during each PCA calculation;
[0077] The application scenario of the embodiment of the present invention is to realize the real-time prediction of the biochemical oxygen demand by establishing the biochemical oxygen demand BOD soft sensor model based on the RBF neural network.
[0078] The inventor found through research that the prediction model ...
Embodiment 2
[0122] On the basis of Embodiment 1, the specific manner of training and learning of the BOD soft sensor model in the embodiment of the present invention may include:
[0123] RBF neural network error e after the tth PCA calculation t (k) expression is:
[0124]
[0125] where, where, q is the number of neurons in the input layer, y td (k) is the expected output of the RBF neural network at time k after the t-th PCA calculation, y t (k) is the actual output of the RBF neural network at time k after the tth PCA calculation;
[0126] Using the gradient descent algorithm to train and learn the BOD soft sensor model, set e d is the ideal error, when e t (k)d , stop the adjustment.
[0127] Under the condition of working condition 1 and working condition 2, the test data of variable input BOD soft sensor model based on adaptive RBF neural network are shown in Table 2 and Table 3 respectively;
[0128] Table 2:
[0129]
[0130]
[0131] table 3:
[0132]
[0133...
Embodiment 3
[0136] On the other side of the embodiment of the present invention, a biochemical oxygen demand soft measuring device is also provided, Figure 4 It shows a schematic structural diagram of the biochemical oxygen demand soft measurement device provided by the embodiment of the present invention, and the biochemical oxygen demand soft measurement device is compatible with Figure 1 to Figure 3 The device corresponding to the biochemical oxygen demand soft-sensing method described in the corresponding embodiment, that is, realized by means of a virtual device Figure 1 to Figure 3 In the biochemical oxygen demand soft measurement method in any corresponding embodiment, each virtual module constituting the biochemical oxygen demand soft measurement device may be executed by electronic equipment, such as network equipment, terminal equipment, or server. Specifically, the biochemical oxygen demand soft measuring device in the embodiment of the present invention includes:
[0137] ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


