Вы находитесь на странице: 1из 6


An injection moulding machine is a machine for manufacturing products by injection moulding process. The analysis of power consumption data during various operational states is very important in industrial point of view.

The main objective is to classify the different operating states from real time data signals based on the smoothened energy consumption data. Initially the signal is smoothened by using advanced digital signal processing techniques like kalman, moving average filter and Savitzky-Golay filter. Of these Moving average filter is found to be efficient as it reduce random noise.

This smoothened signal can be accurately classified into different operational states by using pattern recognition algorithms like support vector machine, Fuzzy C means , Quality threshold and K-Means. K-means clustering is initially trained using filtered power consumption data signals. The different operational states in smoothened energy consumption data can be accurately classified by this trained K-means Clustering.


K-Means clustering classifies the given input data. Technically speaking, K-Means clustering and regression prediction tool that uses machine learning theory to maximize predictive accuracy . K means algorithm performs classification by construction an N-dimensional hyperplane that optimally separates the data into various categories. The power consumption data has six operating states. They are switch off, warm up, idle, start up,production, hydraulic pump/heater. The data obtained during these operating states are plotted and a hyperplane is drawn to differentiate them. K-Means clustering classifies the data signal into six operating states. This iterative partitioning
minimizes the sum, over all clusters, of the within-cluster sums of point-to-cluster-centroid distances. [IDX,C] = kmeans(X,k) returns the k cluster centroid locations in the k-by-p matrix C. [IDX,C,sumd] = kmeans(X,k)

returns the within-cluster sums of point-to-centroid distances in the 1returns distances from each point to every centroid in the n-by-

by-k vector sumd.

[IDX,C,sumd,D] = kmeans(X,k) k matrix D.

K-means is one of the simplest unsupervised learning algorithms that solve the well known clustering problem. The procedure follows a simple and easy way to classify a given data set through a certain number of clusters fixed a priori. The main idea is to define k centroids, one for each cluster. These centroids should be placed in a cunning way because of different location causes different result. So, the better choice is to place them as much as possible far away from each other. The next step is to take each point belonging to a given data set and associate it to the nearest centroid. When no point is pending, the first step is completed and an early group age is done. At this point we need to re-calculate k new centroids of the clusters resulting from the previous step. After we have these k new centroids, a new binding has to be done between the same data set points and the nearest new centroid. A loop has been generated. As a result of this loop we may notice that the k centroids change their location step by step until no more changes are done. In other words centroids do not move any more.

1. Place K points into the space represented by the objects that are being clustered. These points represent initial group centroids. 2. Assign each object to the group that has the closest centroid. 3. When all objects have been assigned, recalculate the positions of the K centroids. 4. Repeat Steps 2 and 3 until the centroids no longer move. This produces a separation of the objects into groups from which the metric to be minimized can be calculated.

clc; clear; %LOADING THE INPUT SIGNAL load inputdata.mat %FILTER THE SIGNAL USING MOVING AVERAGE FILTER fsignal1 = fastsmooth(data1,11); fsignal2 = fastsmooth(data2,11) fsignal3 = fastsmooth(data3,11); %PLOT THE FILTERED SIGNAL DATA1 or DATA2 or DATA3 plot(fsignal1) hold on %CLASSIFY THE FILTERED SIGNAL INTO SIX DIFFERENT OPERATIONAL STATES %SO,WE TOOK 6 cluster=kmeans(fsignal,6) plot(cluster) %IF WE WANT TO CLASSIFY INTO TWO DIFFERENT STATES,WE CAN TAKE %KMEANS(FSIGNAL,2) cluster1=kmeans(fsignal3,2) plot(cluster1)




Thus we implemented the mathematical model of the system in order to detect the signal accurately by using advance signal processing technique. We have documented both input and output plots. The moving average is the most common filter in Digital signal processing, mainly because it is the easiest digital filter to understand and use. In spite of its simplicity, the moving average filter is optimal for a common task: reducing random noise while retaining a sharp step response. This makes it the premier filter for time domain encoded signals. However, the moving average is the worst filter for frequency domain encoded signals, with little ability to separate one band of frequencies from another. Relatives of the moving average filter include the Gaussian, Blackman, and multiplepass moving average. These have slightly better performance in the frequency domain, at the expense of increased computation time.

we classified given three filtered signals into different operational states using K-MEAN Clustering method. By using this K-MEAN Clustering , we calculated six centroid points because there are six different operation states. Based on centroid ,we classified the given filtered signal into switch off, startup, idle, production, warm up and heating etc.